This page intentionally left blank

Oxford University P ress

Oxf or d New York
At hens Auckl and Bangkok Bogota Bombay
Buenos Ai r es Cal cut t a Cape Town Dar es Salaam Delhi
.Florence Hong Kong I s t a n b u l Kar achi
Kua kt Lumpur Madr as Madr i d Mel bour ne
Mexi co Ci t y Nai r obi Pari s Si ngapore
Tai pei Tokyo Toront o
and associ at ed c ompa ni es i n
Ber l i n Ibadan
Copyright 1996 by Stephen Braun.
Publ i shed by Oxf or d U ni v e r s i t y Press, hi e.
198 Madi son Av enue, New York, New York 10016
Oxf or d is a regi st ered t r ademar k of Oxf or d U ni v er s i t y Press
All ri ght s r eser v ed. No part of t hi s publ i c a t i on may be reproduced, stored in a r et r i ev al syst em,
or t r a n s mi t t e d , i n any f or m or means, el ect r oni c, mechani cal , phot ocopyi ng, recordi ng,
or o t h e r wi s e , wi t hout t he pri or per mi s s i on of Oxf or d U ni v er s i t y Press,
Li br ar y ot Congress Cat al ogi ng- i n- Publ i cat i on Data
Br aun, St ephen.
Buzz : the .science; and l ore of al cohol and c a f f e i ne /
by St ephen Br a u n ,
p. em. I n c l u d e s bi bl i ogr aphi cal references and i ndex.
ISBN 0-19-509289-9
I . Al c ohol Popul a r wor ks. 2. Ca f f e i n e - - Po pu l a r works.
} . Ti t l e.
QP80LAW3 1996
61 $' .7828dc20 95-47790
The a u t h o r and publ i s her t hank t he f ol l owi ng f or per mi ssi on t o r e pnni speci fi ed ma t er i a l :
From One Fish, Two Fish, Red Fish, Blue Fish by Dr. Seuss. T'M and copyr i ght I960 and
renewed 1988 by Dr. Seuss Ent er pr i s es , L.P, Reprinted by per mi s s i on of Random House, I nc.
Fxcerpt cd wi t h per mi s s i on ot Scr i bncr , a Di v i si on of Si mon & Schuster, f r om A Moveable
Feast by Er nes t Hemi ngway. Copyr i ght 1964 by Mar y Hemi ngway. Copyr i ght renewed 1992
by J ohn I I . Hemi ngway, Pa t r i c k Hemi ngway, and Crcgory Hemi ngway.
9 8 7 6 5 4 3 2 1
Pr i nt ed in the Uni t ed States of Ameri ca
on aci d free paper
To the memory of
Robert Arnold Braun
Shaman, Scientist, Father
This page intentionally left blank
The idea for this book germinated duri ng a fellowship in neu-
robiology at the Marine Biological Laboratory in Woods Hole,
Massachuset t s. There I was introduced to a radical new un-
derst andi ng of the br ai n and the way it worksand there, too,
I saw clearly for the first time how substances such as alcohol
and caf f ei ne could af f ect the machi ner y of the mi nd. I am
indebted to Irwin Lev i t an, the scientist in charge of the lab in
which I was a st udent , for his pat i ent explanations, s t eadf as t
encour agement , and i nf ect i ous ent husi asm for ncurosci encc.
To the MBL I owe t hanks as well, for of f er i ng science j our nal -
v i i i / Acknowledgments
ists such a uni que oppor t uni t y to experience science by doing
i t , rat her t han si mpl y report i ng i t .
Buzz has benef i t ed t r emendousl y f r om the keen-eyed review
of many sci ent i st s, foremost among them is Steven N. Treist-
man, who car ef ul l y and t hought f ul l y read the ent i re manuscri pt
and who was an unfl aggi ng s uppor t er of the pr oj ect f r om day
one. In addi t i on, t he following sci ent i st s took t i me f r om busy
schedules to read selected chapt er s or chapt er port i ons: Gary
Kapl an, Thomas Dunwi ddi e, Barry Green, Robert Greene, and
Annet t e Rossingnol. Other sci ent i st s helped by answeri ng of t en
lengthy lists of quest i ons and by prov i di ng copies of papers on
rel ev ant s ubj ect s . My thanks to Robert D. Blitzer, Joseph
Brand, James Brundagc, Michael Charncss, John Daly, Da-
vid Lovinger, Quent i n Regcstein, Forrest Weight, and Mark
Whi t ehead.
For the excel l ent mol ecul ar models i ncl uded in the book, I
t ha nk Joe Gambino of the Uni v er s i t y of Massachuset t s Medical
Cent er . The superb line drawings of ion channel s and the neu-
ronal synapse are the work of Ann Bliss Pilchcr.
I am gr at ef ul to Howie Fraz in and Lynn Prowitt for def t l y
finding the conf usi ng par t s and helping me cl ar i f y them. My
editor at Oxford Uni v er s i t y Press, Ki rk Jensen, provided inval-
uabl e gui dance and repeatedly led me to the correct voice for
the book. I am al so most gr a t ef ul to copy editor Gail Weiss,
whose many v al uabl e suggestions improved Buzz considerably.
Finally, my hear t f el t t hanks to Mary Anna Towler for teach-
ing me to wri t e; to Steve Pilcher, Tim Braun, and Doug Beyers
for i nv al uabl e lessons in percept ual r el at i v i t y; to Oralec Stiles
for the f abul ous quote; and, above all, to my wi fe, Susan Red-
ditt, for pat i ence far beyond the call of dut y. Were it not for
her loving support in mat t er s great and small, this book would
not have been possible.
I nt r oduct i on, 3
1. Alcohol 101, 8
2. Down the Hat ch, 20
3. Your Brai n on Alcohol, 38
x / Contents
4. Sex, Snores, and Stomach Aches, 61
5. Demon Rum, 88
6. The Eyelids of Bodhi dharma, 107
7. A Quicker Geni us, 123
8. The Body, Wired, 137
9. Hooked, 162
10. Bet t er Living Through Chemi st ry, 182
Postscript, 193
References and Suggested Reading, 197
Index, 206
This page intentionally left blank
Aristotle, in The Problemata, posed the following questions:
Why are the drunken more easily moved to tears? Why is it
that to those who are very drunk everyt hi ng seems to revolve
in a circle? Why is it that those who are drunk are incapable
of having sexual i nt ercourse? Aristotle considered the brain
not hi ng more than a radi at or for cooling blood, so it's not sur-
pising t hat he couldn' t answer these questions. He and others
at t ri but ed the i nt oxi cat i ng powers of wine and beer to myste-
rious "spirits" of i nebri at i on.
In a similar vein, sev ent eent h-cent ury doctors puzzled over
4 / Buzz
t he st i mul at i ng ef f ec t s of cof f ee and tea. Some argued that the
beverages cont ai ned "cold and moist" essences t hat altered the
bal ance of the body' s f our v i t al fluids, or humor s. Others, how-
ever, t hought t hat cof f ee and tea should be cl assi fi ed as "warm
and dry" in the humor al spect rum. The issue was never re-
solved, t hough it exercised some of the era' s best medical
mi nds for decades.
Alcohol and caf f ci cne t oday are the world' s most widely con-
sumed mind al t eri ng substances, and people are as curi ous as
ever about what t hey are and how t hey work. Like Ari st ot l e,
they wonder why, when they' re dr unk, they see things spi nni ng
and why alcohol can deaden sexual response. In addi t i on, peo-
ple have new quest i ons, ari si ng as oft en from media reports of
s ci ent i f i c st udi es as f r om popul ar myt h. Do women become
i nt oxi cat ed more easily t han men? Does c a f f ei ne worsen pre-
mens t r ual sympt oms? Is al cohol i sm a genetic di sease? Is caf -
f ei ne bad for you or i sn' t it? Does alcohol real l y ki l l brai n cells?
Can caf f ei ne help you lose wei ght ?
The answers to such quest i ons elude even sophi st i cat ed con-
sumer st hose who know t hei r cabernet sauv i gnon from t hei r
sauv i gnon bl anc, and t hei r Kenya AA f r om t hei r Aged Sumat r a.
The reason is simple: u n t i l v ery recent l y, nobody has been able
t o answer these quest i ons. It' s not t ha t alcohol and caf f ei ne
are t er r i bl y complex or di f f i c ul t t o under s t and. In f act , t hey arc
rat her simple mol ecul es, the st r uct ur e of whi ch has long been
known. The problem is t hat the target of those mol ecul es, the
human br ai n, is t erri bl y complex and di f f i c ul t t o under st and.
Progress in l earni ng how alcohol and caf f ei ne work has had to
wait for new knowledge of how the brai n works.
Fort unat el y, in the past two decades the sci ence of the
br ai nneur osei cnechas blossomed. New i nv est i gat i v e tech-
ni ques hav e opened up the black box of t he brai n and have
introduction I 5
begun to shed light on its inner workings. Out of t hi s new
insight has come a radi cal l y unproved under st andi ng of how
alcohol and caf f ei ne work. Much of this i nformat i on is so new
that it is known only to cert ai n research scientists and people
who read sci ent i fi c j our nal s. And on the rare occasion that new
findings have made it f r om the lab bench to the corner bar,
the message of t en arri v es garbled.
For example, an early study suggesting that alcoholics seem
to have fewer brai n cells t han nonalcoholics was wi del y publ i -
cized and cont r i but ed to the idea t hat alcohol kills brain cells
( Har per and Krill 1990). Later studies clearly disproved t hi s
idea, but the notion remai ns. Anot her common idea, based on
a s ci ent i f i c paper i nv est i gat i ng the metabolic ef f ect s of caf f ei ne,
is that dri nki ng cof f ee helps bur n off fat (Costill et al. 1978).
There is a grain of t r ut h to this idea, but as we'll sec, it' s a
very small grain indeed.
Advances in ncuroscience have also t urned some long-
standing ideas on t hei r head. For instance, many people learn
in high school that alcohol is a depressant a kind of chemical
sledgehammer for the mi nd. Alcohol is act ual l y much more
complex. It produces ef f ect s t hat mimic those of many other
drugs, such as opium, cocaine, Valium, and ether. The resul t
is an i nt oxi cat i on far more dynamic and complicated t han most
people realize. Caf f ei ne is similarly mi sunder st ood. It is not a
direct st i mul ant , like a shot of adrenaline. Instead, it works
i ndi rect l y by i nt er f er i ng with one of the brai n' s main chemical
"brakes." Like a car wi t h a sticky brake pedal, the brai n speeds
up because it can' t slow clown.
Some of the new findings, though, support previ ousl y un-
subst ant i at ed fol k wisdom about alcohol and caf f ei ne. Moder-
ate doses of alcohol, for instance, have been shown to have
some heal t h benefi t s, such as reducing the risk of cardi ov as-
6 / Buzz
cul ar di sease. And, as ma n y at hl et es have long suspect ed, caf -
fei ne has been f ound to si gni f i cant l y boost per f or mance in a
number of sports, such as r unni ng and bicycling.
In short, recent sci ent i fi c progress has r ev ol ut i oni z ed the un-
der st andi ng of alcohol and caf f ei ne. This rev ol ut i on has not yet
been recognized by the public, despi t e the fact that these two
subst ances are consumed dai l y by a ma j or i t y of huma n beings
on the pl anet . Caf f ei ne is the most popul ar drug on eart h (Gil-
bert 1984). It is cont ai ned in tea, cof f ee, cocoa, chocolate prod-
uct s, many sof t dri nks, and more t han 2,000 nonprcscri pt i on
drugs (Robson 1992). In t he Uni t ed States, roughly 80 percent
of adul t s consume c a f f ei ne in one f or m or a not her every day
( St r ai n ct al. 1994).
In the Uni t ed States, alcohol is second only to caf f ei ne as
t he drug of choice. Sixty-eight percent of Ameri can men and
about 47 percent of Amer i can woman say t hey dr i nk alcoholic
beverages at l east occasi onal l y ( Depar t ment of Health and Hu-
man Services 1993) . In other count r i es where the r at e of smok-
ing is rel at i vel y high, ni c ot i ne t akes second place, nudgi ng
alcohol to t he number - t hr ee posi t i on. Thus, with the except i on
of wat er , all of the most popul ar beverages on eart h cont ai n
ei t her c a f f ei ne or alcohol.
This book is about how these two phenomenal l y popul ar
subst ances work: how t hey a f f ec t t he br ai n and t he body, arid
how t hey are cur r ent l y underst ood in light of recent sci ent i f i c
advances. This i nf or mat i on is laced t hr oughout with cul t ur al
hi st or y and personal st ori es. Mi khai l Gorbachev , David Let-
t er man, I
1
'. Scott Fitzgerald, William Shakespeare, Buddhi st
monks, and Arabi an goat herds make cameo appear ances. We'll
hear what j ohann Sebastian Bach, Teddy Roosevelt, and J ohn
Steinbeck hav e to say about alcohol and c a f f e i n e . We'll meet
a champion swi mmer suspended from compet i t i on for two
years because she ingested too much caf f ei ne, and a group of
Introduction I 7
college st udent s who helped reveal how alcohol af f ect s sexual
response.
Human stories such as these are inevitable because alcohol
and caf f ei ne are so deeply and i next ri cabl y woven into the fab-
ric of daily l i f e for most of the eart h' s population. They are
fami l i ar drugsas close and comfortable as a cup of coffee or
a can of beer. And yet t hey and the buz z they produce remain
for most people j us t as myst eri ous and unpredi ct abl e as the
spirits they were once thought to be.
The Green Dragon
In the mid-1980s, citizens of the Soviet Uni on were faced with
a nat i onwi de shortage of sugar so serious t hat this basic f ood-
s t uf f was rationed l i ke gasoline. The cause of the shortage lay
not in the usual cul pri t s of i nef f i ci ent st at e-run product i on,
i nadequat e i mport s, or dilapidated di st ri but i on systems. No, it
was zelyony zmeithe green dragon. That was the ni ckname
given to Mi khai l Gorbachev' s crusade to wean his count r ymen
f r om t hei r nat i onal dri nk: v odka. The name r ef er r ed to the coils
of of t en greenish copper pipes used in the t housands of home
Alcohol 1 0 1 / 9
distilleries built to supply what the government was t ryi ng to
l i mi t .
The nat i on' s sugar supply was di sappeari ng into these secret,
jury-rigged stills. Of course, Soviet moonshiners also made
booze f r om potatoes, corn, wheat, barley, and other st archy
mat eri al s. But these raw i ngredi ent s complicated the process.
To be us ef ul for f er ment at i on, starch molecules first have to
be broken down by enzymes i nt o the i ndi v i dual sugar mole-
cules from which they are made. It is much simpler and neat-
erthough not necessarily cheapersimply to start with raw
sugar. Soviet bootleggers pouring bags of sugar into their home
f er ment at i on vat s doubtless do not dwell on the details of the
t r ansf or mat i on they shepherd: the conversion of ordi nary table
sugar into a powerful mi nd- al t er i ng drug. But it is a remarkable
f eat of alchemy indeed.
For most of human hi st ory, this t r ans f or mat i on was a deep
myst ery, even though beer- and wine-making are among civil-
i z at i on' s earliest professi ons. Recent anal ysi s of a yellowish sub-
stance f ound in the bottom of clay j ar s unear t hed in Iran
confi rms that the Sumeri ans were accomplished brewers 5,500
years ago (Wilford 1992). And brewing wasn' t a t r i v i al enter-
pri se. One of the most common pictographs in Sumeri an rui ns
is the sign for beerhatch mar ks t hat represent the crisscross
pat t erns dug into the bottom of beer vessels.
The Sumeri ans are the most anci ent users of alcohol that
we know about. But the discovery of f er ment at i on was a world-
wide phenomenon. Where there was sugar, humans learned
ways to encourage its conversion into alcohol by ferment at i on.
But although the art of f er ment at i on is anci ent , the science is
not. As ci vi l i z at i on flourished over the centuries, people made
beer and wine with only the crudest under st andi ng of what
they were doing. It wasn' t unt i l qui t e late in the game that
anyone had a clue about what alcohol was or how it was pro-
10 I Buzz
duced. Underst andi ng these two things is the first step in un-
derstanding such larger issues as intoxication, hangovers, and
addictiontopics we'll explore in later chapt ers.
Here we'll simply get to know alcohol for what it really is.
Portrait
If you could examine beer, wine, or liquors under a super-
powerful microscope, you would see a wild j umbl e of mole-
cules bashing into one another in a confusi ng t umul t . If you
zeroed in on a single molecule of alcohol, here' s what you
would see:
Ethanol
This is what people get when they order a dri nk at a bar
a type of alcohol called ethanol. You can see that its nine con-
st i t uent atoms nestle against one another, f or mi ng a l umpy
particle that, in this part i cul ar ori ent at i on, bears more than a
passing resemblance to an exceptionally pudgy dog.
Ethanol is made up of two carbon atoms (shown in black),
an oxygen atom (in gray), and six hydrogen atoms (in whi t e).
Notice the "head" of the dog. It' s the oxygen-hydrogen group
on the upper right. Any molecule with this group attached to
a carbon atom is called an alcohol. Since this is a f ai r l y com-
mon arrangement in molecules, there are lots of alcohols in the
world. Glycol is the alcohol in ant i freez e, and cholesterol is a
complicated type of alcohol vital for many bodily funct i ons
Alcohol 1 0 1 I I I
(and in excess, helps to clog blood v essel s). Ethanol is simply
the most f amous member of a very large f ami l y of alcohols,
though it' s not the only alcohol that causes i nebri at i on. You
also can get drunk on ethanol' s si mpl er cousin, methanol,
which can be produced by the f er ment at i on of wood and thus
has the common name wood alcohol:
Methanol
Although cheap to produce, methanol has a rat her serious
drawback as an i nt oxi cant : it is broken down in the body by
an enz yme f ound in part i cul ar abundance in the ret i nas of the
eyes. This enzyme converts methanol into a closely related
molecule: formal dehyde. As anyone who has dissected a pre-
served frog knows, formal dehyde is not something you
part i cul arl y want inside your eyeballs. But that is exactly what
happens when people dri nk methanol. The enzyme in their
retinas converts methanol to formal dehyde, causing permanent
blindness. Ethanol' s ext ra carbon atom gives it a shape j ust
di f f er ent enough from that of methanol to be unaf f ect ed by
that enzyme in our retinas. We t hus can dri nk freely with the
knowledge that although we may not be clearheaded the next
morning, we will at least be clear-eyed enough to find the med-
icine cabinet.
Another of ethanol' s interesting charact eri st i cs is its di mi n-
utive st at ur e. Compare ethanol with some other molecules
that produce i nt ri gui ng resul t s when i nt roduced into human
brains:
Ethanol Morphine Tetrahydrocannabinol (THC)
Morphi ne and TIIC are obviously l arger and more complex
than ethanol. But even these drugs are t i ny compared with the
molecules common in the human body. Here, again drawn to
scale, is ethanol and hemoglobin, the molecule t hat carries oxy-
gen in red blood cells:
Ethanol Hemoglobin
Alcohol 101 I 13
Most of the molecules af f ect ed by alcoholincluding those
involved with intoxicationbelong to t hi s j umbo-si z e class of
molecules.
Of course, ethanol isn' t j us t relatively small. It' s i mport ant
to appreciate j ust how many ethanol molecules are f ound in
a standard dr i nkt hat is, a 12-ounce can of beer, a 5-ounce
glass of wine, or a 1.5-ounce shot of l i quor. Each of these
dri nks typically cont ai ns a hal f-ounce of pure ethanol, or ap-
proximately 200 quintillion ethanol molecules, each of which
has the ability to di srupt some part of your body's cellular ma-
chinery.
Size isn' t the only i nt erest i ng aspect of ethanol' s st ruct ure.
Ethanol molecules also carry a small electric charge that is cru-
cial to its behavior in the body. The oxygen atom in the "head"
of the molecule makes that region slightly negative. This
charge meshes nicely with the slight positive charge on one
side of water molecules. It looks something like this:
Ethanol Water
Without the electric charge of its "head" region, ethanol would
separate f r om water like oil. Not only would this make mi xi ng
a gin and tonic an exercise in f r ust r at i on, but it would make
it di f f i cul t for ethanol to penet rat e our water-filled bodies.
Oddly enough, however, ethanol is also soluble in oils and f at s.
That' s because the "tail" end of the molecule is not signifi-
cant l y charged. This region would prefer, electrically speaking,
not to associate with water, and so it meshes easily with fat
and oil molecules that would also rather avoid water. Ethanol
! 4 / Buzz
is t hus a powerful solvent t hat can roam f r eel y t hr oughout the
body. Because it dissolves easily in wat er, it is rapi dl y absorbed
from the digestive t ract and mi xes easily with blood. Because
it also dissolves in f at s , it f r eel y passes t hr ough cell membranes,
whi ch are basically double-walled bubbles of f at .
The overall st r uct ur al shape of ethanol is i mpor t ant as well.
A key idea in chemi st ry is t hat the physical shape of a molecule
can be critical to its behavi or in the world. The knobby pro-
t rusi ons and i nf ol di ng pockets of most molecules keep them
from t ouchi ng in a si gni f i cant manner. The bumps usually get
in t he way. But sometimes one molecule' s knob will happen to
fit j us t per f ect l y i nt o anot her molecule' s pocket like a key in a
lock. When t hat occurs, all sorts of i nt erest i ng things can hap-
pen. The two molecules may war p or twist slightly under the
i nf l uence of the connection. Normal l y stable molecules may
suddenl y become unst abl e or split al t oget her.
As we' ll see, t hese physi cal shape-to-shape i nt eract i ons un-
derl i e many of ct hanol ' s ef f ect s in our br ai ns and bodies. The
way t hat ct hanol ' s shape mat ches the par t i cul ar shape of cer-
t ai n key pr ot ei ns and ot her molecules in the br ai n has every-
t hi ng to do wi t h why we get i nt oxi cat ed.
Yeast Trash
Now we know that the "spirits" in wine, beer, and l i quor are
real l y quadr i l l i ons of et hanol molecules. But exactly where do
those mol ecul es come f r om? Consi deri ng t hat hur nans have
been maki ng alcohol for at least 5, 500 years, it has taken a
sur pr i si ngl y long time to figure t hi s out. Under st andi ng alco-
hol' s origins will shed light on some mi nor puz z l es about how
i t behav es in the body.
Soviet bootleggers know part of the answer . Alcohol comes
f r om sugar. They also know t hat t hi s t r ans f or mat i on isn' t spon-
Alcohol 101 I 15
t aneousi t requi res the assistance of one-celled fungi called
yeast. This much has been known for centuries. But it wasn' t
unt i l 1939 that the f ul l det ai l s of the st ory were fi nal l y worked
out by a scientist named Embden O. Meyerhof. Meyerhof was
t r yi ng to figure out how sugar is metabolized in our bodies;
when he solved t hi s problem, he also had the answer to the
age-old quest i on of where alcohol comes f r om.
The process st art s with glucose, which is the sugar both hu-
mans and yeast use to power t hei r bodies. Other sugars, such
as the sucrose (table sugar ) used by the Soviet home distillers,
must be converted to glucose before t hi ngs get going, chemi-
cally speaking. Like humans, yeast cells pr ef er to burn t hei r
glucose with oxygen to produce energy. But yeast cells some-
times find themselves in si t uat i ons where oxygen is scarce
for i nst ance, when they are trapped in the bottom of huge v at s
of grape j ui ce. In such ci rcumst ances, they manage to carry on
by using backup metabolic machi nery designed to bur n glucose
without oxygen. This anaerobic system is much less ef f i ci ent
than the pri mary, oxygen-using system. Instead of gradual l y
(and completely) breaking down the glucose molecule with ox-
ygen to release lots of energy, the anaerobic system simply
splits the glucose in two, which resul t s in a rel at i vel y feeble
amount of energyj ust enough to sust ai n mi ni mum l i f e f unc-
tions unt i l oxygen r et ur ns.
This spl i t t i ng isn' t accomplished in a single whack, like a log
split with an ax. One of Meyer hof ' s contributions was the dis-
covery t hat it takes ten separate steps to split glucose wi t hout
oxygen. A series of ten separate enzymes is used to twist and
pick apart the original glucose molecule unt i l it can easily be
cleaved. The process resembles nothing as much as a mol ecul ar
disassembly line .in which the glucose molecule is worked on
by one enzyme and then is passed on to the next and so on.
The details of t hat process are i nt erest i ng in their own right,
16 / Buzz
but all we' re real l y concerned with here are those two shards
remai ni ng a f t er t he glucose is finally spl i t . Those shards are
molecules of et hanol .
The bi r t h of alcohol via t hi s i nef f i ci ent spl i t t i ng of glucose
has one very sal i ent consequence for humans : most of the
chemi cal energy of the ori gi nal glucose molecule remai ns
bound up in the ethanol f r agment s. That energy equal s calo-
ri cs: about seven per gramwhi ch works out to about a hun-
dred calories in a st andar d dr i nk from t he alcohol alone.
Alcohol, in ot her words, is no diet dr i nk.
Alcohol' s ori gi ns also expl ai n some f act s about the alcohol
cont ent of some common dr i nks. Yeast cells struggling to sur -
vive under suf f ocat i ng conditions qui ckl y excret e t he et hanol
f r agment s because t hey are basically poisonous. Ethanol inter-
f er es with many of the react i ons vi t al to the l i f e of a cell. As a
r esul t , yeast s excrete ethanol, which slowly bui l ds up in the
sur r oundi ng l i qui dexact l y where the brewer or v i nt ner want s
it. Given an adequat e amount of glucose, the ethanol cont ent
of a f er ment i ng l i qui d rises unt i l it reaches about 12 percent .
At t hi s poi nt , it st art s to back up inside the yeast cells because
it can no longer di f f us e across t he cell wall. Unable to dispose
of the poisonous waste, the yeasts shut down and become dor-
mant . All act i v i t y stops, i ncl udi ng the pr oduct i on of new eth-
anol . This is the reason that most table wines have roughly a
12 percent alcohol content: t hat ' s as high as it can go before
the yeasts t hrow in the towel. Some wines can achieve slightly
higher v al ues if t hey are unusual l y rich in glucose, but the only
way to get si gni fi cant l y hi gher et hanol levels is by di st i l l at i on
the gent l e boiling of a l i qui d in a scaled cont ai ner.
Di st i l l at i on, by the way, is possible only because et hanol
molecules happen to ev aporat e more qui ckl y t han wat er mol-
ecul es f r om a l i qui d mi xt ur e such as wine or beer. Water mol-
ecules carry an el ect ri c charge and t end to stick to one anot her .
Alcohol 101 I 17
Ethanol, however, is only weakly chargedand only at one
end. That leaves ethanol molecules f r ee to escape at t emper-
at ur es lower t han the boiling poi nt of wat er. If these f r ee eth-
anol molecules are capt ured (by condensation eoils, for
i ns t ance) , any desi red concent r at i on up to 96 percent can be
achieved. A 100 percent ethanol solution is impossible to
achieve wi t h or di nar y di st i l l at i on t echni ques because some wa-
ter molecules unav oi dabl y escape along with the ethanol. Com-
merci al l y pur e et hanol is produced using a v ar i et y of chemical
react i ons t o el i mi nat e t he wat er .
Of course, the ethanol concent rat i on of beverages is mea-
sured in two ways: by percent alcohol and by "proof." The t erm
"proof" dates to the seventeenth cent ury when various means
were devised for checking, or "proving," that a beverage had
the alcohol content its label cl ai med. The proof number is j us t
about double the percent age by v ol ume of alcohol. A wine with
12 percent alcohol, in other words, is roughly 24 proof.
The Human Connection
We' ve now seen t hat et hanol is molecular garbage made by
yeasts when they bur n glucose under the duress of s uf f ocat i on.
As remote as t hi s s i t u a t i o n is f r om dai l y l i f e, t here is, in f act ,
an i nt er est i ng connect i on between yeast and humans. It t ur ns
out t hat we have al most exactly the same cellular machi nery
in our bodies t hat yeasts doand for the same reason.
Like yeast , the cells in our bodies usual l y bum glucose wi t h
oxygen because it rel eases so much energy. But even for highly
mobile huma ns , oxygen isn' t always avai l abl e. In f act , it' s ex-
act l y those par t s of the body t hat are used most vigorously that
may face an oxygen s hor t f al l . In st r enuous runni ng, for exam-
ple, cert ai n muscl es use oxygen more quickly t han it can be
replenished by the blood, l eadi ng to a localized condition rei n-
18 / Buzz
i ni sccnt of t hat faced by yeast: in the bot t om of f er ment at i on
v at s. At such t i mes, muscle cells fal l back on the same i nef f i -
cient anaerobic machi ner y used by yeast machi ner y we in-
heri t ed from our single-celled ancest ors.
Act ual l y, t he ent i r e mol ecul ar appar at us i s retained except
for a small, but i mpor t ant , al t er at i on. Since yeasts arc t i ny or-
gani sms, they can easi l y di spose of t hei r et hanol t rash. Hu-
mans, however, bei ng rel at i v el y enormous, must rely on a
complicated wast e- el i mi nat i on system. If we produced ethanol
as a wast e product of met abol i sm, as do yeast cells, it would
ci r cul at e in our blood and we'd end up ri p-roari ng dr unk when-
ever we exerci sed har d. This obvi ousl y wouldn' t be a very adap-
t i v e si t uat i on f r om an ev ol ut i onar y st andpoi nt . It' s not
surpri si ng, t her ef or e, t hat i n humans and other ani mal s t he
process of anaerobic metabolism is slightly altered to avoid the
product i on of et hanol . In humans , the last of the ten steps
requi red to split glucose is changed so t hat the two r es ul t i ng
f r agment s are molecules of l act i c acid, not ethanol. Too much
lactic acid, of course, can cause muscle soreness and f at i gue,
but at l east i t l ef t our f or ebear s sober whi l e t hey fl ed f r om
saber-t oot hed tigers and t hei r ilk.
Molecular Spi ri ts
This chapt er has i nt roduced a new way of looking at alcoholic
beverages. The "spirits" in these dr i nks are act ual l y t r i l l i ons
upon t r i l l i ons of i ndi v i dual ethanol molecules. These molecules
are small and knobby and have some pecul i ar at t r i but es, such
as the abi l i t y to dissolve in both wat er and f at s char act er i s t i cs
that are key to t hei r abi l i t y to breach the body's normal de-
f ens es . We've also seen t hat alcohol is act ual l y a type of ex-
crement produced by s uf f ocat i ng yeast . It exists as a met abol i c
Alcohol 1 0 1 / 1 9
waste product and would be produced by humans as well ex-
cept t hat nat ur e apparent l y didn' t select for creat ures with this
ability.
All of t hi s adds up to a rat her uncommon perspective on
alcohol. But it is not yet a par t i cul ar l y hel pf ul perspective be-
cause as i nt erest i ng as it may be, this new underst andi ng of
alcohol's t rue nat ur e is only half the equat i on of i nt oxi cat i on.
The other hal f is the human brai n.
A Wee Dram
I magine that you find yoursel f floating, mysteriousl y, in a shot
glass containing a drain of a nice scotch whiskyl et's say it's
eighteen-year-ol d Macal l an. You've been shrunk (al ong with
appropriate scuba gear and a powerful flashlight) by a factor of
about a bil l ion to the size of a small mol ecul e. At this scale, a
single cthanol molecule is roughl y the size of a corpul ent Lab-
rador retriever.
As you gl ance around, you see cthanol molecules wiggling in
al l directions, along with si mi l arl y freneti c water molecules and,
here and there, l arger, more globular mol ecul es you can't idcn-
Down the Hatch I 21
t i f y. These are probably the phenols, sugars, t anni ns, and in-
organic compounds t hat give the scotch its amber color, smoky
aroma, and di st i nct i v e t ast e. Many of these molecules seeped
i nt o the i ni t i al l y crystal-clear whisky from the oak casks in
which the l i quor was aged. This part i cul ar whisky is aged in
oak barrel s used prev i ousl y for sherry. That means that some
of the molecules ori gi nat ed in the grapes that went into the
sherry. Some of the other molecules you notice ori gi nat ed in
the smoke f r om t he peat fires used to roast the barley malt
prior to f er ment at i on, and still others are derived from the
barley grai ns themselves.
All in all, the view from inside this shot glass is remarkable.
But you have no time to appreciate it. Suddenly you are caught
in a tidelike surge. The glass is raised ceremoniously (we'll as-
sume in a toast to science r at her than Bacchus) and t hen
t i pped into some anonymous mout h. It is suddenly dark.
The i magi nary voyage you' re about to take will begin our
expl orat i on of how the et hanol we met in the previous chapt er
actually works in the human body. In this chapt er, we'll see
what happens as ethanol moves from the mout h to the stom-
ach, i nt o the blood, and t hen to the l i v er, an organ with a key
role in the experience of i nt oxi cat i on. As you take this t r i p,
you' ll learn about a number of the smaller myst eri es of dri nk-
ing, such as why many people pr ef er drinks "on the rocks" and
why some people are much more sensitive to alcohol than oth-
ers. You' ll also be introduced to some of the f undament al ideas
t hat are key to under st andi ng how both alcohol and caf f ei ne
work in the di f f er ent part s of the body.
Tip of the Tongue
You' re now flowing in a ri v er of molecules across the furrowed
surface of an enormous wri t hi ng slab of muscle: the tongue.
You flick on your light. Although most of the flow is surging
22 / Buzz
toward the back of the t hroat , you and several hundred ot her
molecules arc bei ng dri v en by the cur r ent s down i nt o a deep
canyon.
As you descend, a mushroom-shaped s t r uct ur e looms up out
of the mur k. At your molecular scale, it appears monst rous.
You' re looking at a f ungi f or m papillaone of the 9,000 or so
small bumps on the tongue t hat most people call t ast e buds.
The name "f ungi f or m" r ef er s to the mushroom shape of t hese
burnps, which arc act ual l y j us t support st r uct ur es. Each papilla
cont ai ns between fi ft y and one hundr ed i ndi v i dual taste buds,
most of them located on the underside of the papi l l a, not the
top.
The cur r ent forces you and your company of et hanol mole-
cules up agai nst the papi l l a. Dead ahead, a t ast e bud comes
i nt o view. Not a bud at al l , it looks more like a gaping hole in
the s ur f ace of the papi l l a. Down into the hole you plunge. At
the bottom of t he pi t , you see what looks l i ke a bed of kel p
wav i ng slowly in what is now a r at her viscous mi xt ur e of scotch
and saliva. These are mi cr ov i l ht he fingerlike pr oj ect i ons of
the many i ndi v i dual t ast e recept or cells in each bud. The fin-
gers great l y increase the sur f ace area exposed to the molecule-
l aden cur r ent s washi ng in and out of the t ast e bud' s pore.
As you close in on the mi crov i l l i , they begin to tower over
you l i ke giant pi l l ar s. Now you' r e so close you can sec t i ny
bumps st uddi ng t he ot herwi se r el at i v el y smooth sur f ace of t he
microvilli fingers. A sudden wave shoves you agai nst the pillar
near one of the bumps. You can now see the bump for what
it really is: a s i ngl e, huge pr ot ei n molecule, lodged in the
membr ane of the mi crov i l l us l i ke a t enni s ball stuck in a chain-
l i nk fence. You' re l ooki ng at the part of the "ball" exposed to
the outside world, but the molecule spans the cell membrane
and ext ends back i nt o the body of the mi crovi l l us i t s el f .
Suddenl y the bump moves spasmodically. The ent i r e mole-
Down the Hatch I 23
cule rapi dl y changes shape. One moment it looks like a solid,
protruding mass f r om the cell membrane, and the next mo-
ment, with a qui ck t wi st i ng motion, it opens up, f or mi ng what
looks like a hole or a tunnel. Aiming your light down into it,
you can see right through the membrane to the cell i nt eri or.
From what you can see, the interior is densely packed with a
bewildering array of molecules in all shapes and sizes. With
the t unnel open, you notice thousands of t i ny part i cl es surging
through the breach. These are act ual l y electrically charged at-
omsgenerically called ionsand t hei r surgi ng is one of the
f undament al actions underl yi ng both normal thought and the
altered st at es induced by alcohol and caf f ei ne.
The qui v er i ng, dynamic molecule you' ve been watching is
called an ion channel, because it acts like a gate r egul at i ng the
flow of ions across a cell membrane. Since ions carry electric
charge, the flow changes the electrical env i ronment inside a
cell. Alcohol and caf f ei ne can af f ect the way ion channels open
and close (Fi gures 1 and 2). Like a foot in the door, they can
leave a channel stuck open. Or, alternatively, t hey can make it
harder for a channel to open in the first place. It all depends
on the type of channel involved. In l at er chapt ers, we'll look at
ion channels in the brain as well as other part s of the body and
see what happens when they are i nfl uenced by alcohol or caf-
feine (or bot h).
Here on the tongue, ion channels play a critical role in our
sense of t ast e. The channel you' ve been watching is embedded
in the membrane of a t ast e receptor cell. The part of the chan-
nel that pr ot r udes i nt o the scotch-laden cur r ent s contains cer-
t ai n spots t hat "recognize" the shapes of some of the larger
molecules floating around you. If one of these molecules
sloshes against one of these spots on the receptor, it binds very
bri efl y, causing the entire physical assembly to shi f t into the
open confi gurat i on. Then, in a blink of an eye, the trigger mol-
24 / Buzz
Figure I. Closed ion channel, cross section. (Ann Bliss P ilcher)
cculc detaches and t he ion channel resumes its normal closed
shape. If enough ion channel s open at approxi mat el y the same
t i me (as is happeni ng now f r om the sudden i nf l ux of scotch)
the electrical balance of the ent i r e taste receptor is tipped so
much t hat nearby nerves are st i mul at ed. In t hi s case, those
nerves hre off a message t hat is i nt er pr et ed in the brain as
mi l dl y bi t t er . Perhaps the molecules we've been looking at,
t hen, arc shar p- t ast i ng t anni ns deri v ed, originally, from t he oak
sheaves of the sher r y casks.
This same sequence of event s is happeni ng all over the
tongue. The recept ors for the f o u r basic tastessalt, sour,
sweet, and bi t t er ar e all st i mul at ed to one degree or anot her
by the molecules in the scotch. The end resul t is a complicated
t ast e message sent to the br ai na message made even more
complex by the s i mul t a neous recept i on of signals f r om t he
nose, which has a much more diverse set of recept ors than t he
t ongue.
Down the Hatch I 25
Int erest i ngl y, none of these taste signals is clue to ethanol.
Pure ethanol is v i r t ual l y tasteless, though some people report
that in a weak sol ut i on, ethanol t ast es f ai nt l y sweet. But eth-
anol clearly does i nt er act with a separat e set of nerve recept ors
in our mout h, nose, and esophagus called polymodal pai n fi-
bers. These receptors out number t ast e receptors by about two
to one. They arc very sophisticated nerve cells that respond to
three kinds of s t i mul i : physical pressure, t emper at ur e, and spe-
ci fi c chemicals. When these receptors are ov erst i mul at ed, we
perceive pai n or i r r i t at i on.
From your cur r ent vantage poi nt , you are well positioned to
see how alcohol can get at these pai n fibers. Dr i f t i ng away from
the qui veri ng ion channel, you notice legions of ethanol mol-
Figure 2. Open ion channel, cross section. (Ann Bliss P ilcher)
26 / Buzz
cculcs disappearing across the membranes of nearby cells. Be-
cause t hey' re made of f at , membranes repel most of the
molecules wanderi ng ar ound outside a cell, i ncl udi ng the ubiq-
ui t ous wat er molecules. But this normally i mpenet rabl e bar r i er
is easi l y breached by fat -sol ubl e et hanol molecules, which slip
t hrough l i ke little ghosts. Since you are not fat soluble, you
can' t follow the ethanol into near by cells. But if you could,
you' d see them passing rapidly t hrough several l ayers of cells
to the polymodal pai n fibers lying under neat h.
Exact l y how ethanol st i mul at es these fibers is a myst er y at
the moment . But anyone who has tasted undi l ut ed whiskey,
gin, r um, vodka, or any ot her high-proof dr i nk can at t est t hat
they are, indeed, stimulated by ethanol. Research has shown
that the more these fibers are st i mul at ed by an alcoholic dri nk,
the great er the bur ni ng and i r r i t at i on in the mout h, t hr oat ,
and nose. We experi enee bur ni ng because ethanol somehow
st i mul at es the receptors the same way t hat high t emper at ur e
does. (Ethanol i sn' t the only molecule capable of "tricking"
these receptors. Capsaicin, a molecule produced by many spe-
cies of pepper pl ant s, does the same t hi ng. )
As most dr i nker s know f r om experience, these bur ni ng sen-
sations can be reduced si gni fi cant l y by chilling the l i qui d pri or
to consumpt i on. Chi l l i ng act ual l y serves several purposes. Fi rst ,
cool et hanol molecules have less v i br at i onal energy t han warm
ethanol molecules. Less energy means less impact when the
molecules physically bump i nt o t he mout h and t hroat ' s pai n
receptors. It also reduces the abi l i t y of et hanol to move through
the layers of skin cells to get to the receptors in the first place.
Cooling also makes it harder for et hanol molecules to escape
as vapor. This ef f ect is par t i cul ar l y i mpor t ant when the l i qui d
is cont ai ned in glasses such as brandy s ni f t er s . If the concen-
t r at i on of t rapped ethanol vapor is hi gh, a qui ck s ni f f will prove
Down the Hatch I 27
uncomf or t abl ei f not downright pai nf ul whi ch may i nt er-
fere with one's appreci at i on of the subtler aromas of the dri nk.
Finally, a cold dr i nk directly st i mul at es receptors for coolness.
When these receptors are st i mul at ed, their signals appear to
partially of f s et or at least muddlethe si mul t aneous signals
for heat generated by the fal se st i mul at i on of the nearby po-
lymodal pain fibers.
It's for all these reasons that some people pr ef er ice-cold
beer, chilled wine, and l i quor on the rocks. Others dri nk, at
least in par t , to savor the flavor and aroma of a dr i nkt he
complex but t erscot ch and vanilla flavors of aged bourbon, the
raspberry and cherry notes of a good Zi nfandel , or the nut t y
sweet flavors of a good t awny port, for instance. For these folks,
t emperat ure control is a delicate balance. Chill a dri nk too
much, and the aromas and flavors arc lost. But let it get too
warm, and excessive et hanol is released, i nt er f er i ng with some
of the more pleasant aspects of the dri nk.
Front-Line Defense
By now, as you lie snagged in the l abyri nt hi ne corridors of the
tongue, most of the scotch is long gone. The bulk of the shot
passed i mmedi at el y through the mouth and down the esoph-
agus, where, along the way, ethanol molecules stimulated po-
lymodal pain fibers and generated a gentle burni ng sensation
as it dropped into the stomach. Now, with a swish and a swal-
low, you, too, are sent on your way.
As you fal l into t hi s cavernous organ and look around, you
see that the shot was admi ni st ered before a meal. The stomach
is mostly empt yt he whisky is lying in a shallow pool where
it is now mixed with highly acidic gastric j ui ces. The acid levels
here are 100,000 t i mes higher than those f ound in the blood
&
28 / Buzz
high enough to dest roy most of the l arger molecules in the
whi sky. But ct l i anol mol ecul es, because they are so small and
stable, arc i mmune to acidic dest r uct i on.
As you near the corrugated wall of the stomach, your l i ght
gl i st ens off t he t hi ck l ayer of mucus t hat prot ect s t he stomach
f r om its own aci d. But even t hi s normal l y i mperv i ous mucus is
easi l y breached by ct hanol . As you watch et hanol molecules
di sappear i ng i nt o t he mucus, yon not i ce an i ncreased oozing
f r om many t i ny pi t s i n the l i ni ng i t sel f . Ti ny gl ands at t he
bottom of these pi t s produce stomach acid. Et hanol st i mul at es
t hese gl ands to produce more acid by means t hat arc not yet
underst ood. The i ncr eased acid goes unnot i ced by most people,
but those with ei t her sensitive stomachs or large appetites for
al cohol may exper i ence st omach pai n or i ndi gest i on f r om t he
excess aci d.
As they pass t hrough the st omach wal l , et hanol molecules
also st i mul at e receptor cells sensitive to such vari abl es as acid-
i t y, di st ent i on, and the presence of poisons such as bact er i al
t oxi ns. I f enough of t hes e so-called sent r y receptors arc stim-
ul at ed by et hanol , t hey send an el ect r i cal si gnal up to a small
portion of the brai n called the emetic cent er. A less polite
name would be the "vomit center" because t hi s cl ust er of neu-
rons cont rol s the i nv ol unt ar y muscle mov ement s of retching.
For t unat el y, these sent r y recept ors are not t er r i bl y sensi t i ve t o
et hanol : most people have to dri nk qui t e a bit to trigger them.
But if these cells are al r eady r espondi ng to stomach di st ent i on,
such as f r om a large meal or onc-t oo-many beers, t hen the
added st i mul at i on f r om et hanol can increase the chances for
an unwel come r et ur n of the eveni ng' s ingestion.
Since you are so close to the stomach l i ni ng, you decide to
t ake a swim t hrough the mucus and fol l ow some et hanol mol-
ecul es. Cont r ar y to prev i ous bel i ef s , the stomach is not the
pr i mar y rout e by which et hanol ent er s the blood. Some ab-
Down the Hatch I 29
sorpt i on docs take place, but it is generally i nsi gni fi cant to the
experi ence of i nt oxi cat i on. One reason for this becomes clear
as you slip and sl i t her between mucous molecules. Di rect l y
ahead you see a huge, globular molecule the size of a two-car
garage. It ' s par t of the body' s first l i ne of defense against eth-
anol: a det oxi f yi ng enz yme called alcohol dehydrogcnase. You
watch as an ethanol molecule bumps up against the enzyme.
Nothing happens. The ethanol rebounds, t wi st s, and suddenly
wedges tight into a small crevice in the face of the enz yme.
Inst ant l y, one of the hydrogen atoms on the ethanol molecule
is ri pped o f f . Shorn of its hydrogen atom (hence, dehydroge-
nat ed), the ethanol is released by the enz yme, which is now
ready to take on the next ethanol molecule that happens its
way.
The molecule l ef t behind by t hi s surgery is no longer etha-
nol. It is called acetaldehyde. The removal of that single hy-
drogen atom renders ethanol pharamacologically inactive: you
can' t get dr unk on acetaldehyde. Nonetheless, acetaldehyde is
a very chemically react i v e l i t t l e molecule that, like ethanol, can
seriously i nt er f er e with the mol ecul ar machinery of a healthy
cell. Specifically, it readi l y binds with a wide range of proteins,
which can cri ppl e their normal f unct i oni ng. That' s why your
body is equi pped with a second enzyme in its defense against
alcohol; called aldehyde dchydrogenasc, it is specifically tai-
lored to destroy acet al dehyde. This en/ ymc docs to acetalde-
hyde what alcohol dehyclrogenase does to ethanol: it removes
a hydrogen at om, t hus produci ng acet i c acid, which is har m-
less. In a moment , well take a closer look at t hi s second step
in the body' s detox process. But l et ' s pause a moment and
consider the key player in the first st ept hat garage-size mol-
ecule of alcohol dehyclrogenase.
Why does t hi s enz yme exi st ? Why do we have genes for
bui l di ng an enz yme speci fi cal l y t ai l ored to destroy alcohol?
30 / Buzz
Clearly, t hese genes did not j us t magically appear with the
advent of the human discovery of f er ment at i on. The answer to
these quest i ons lies in our gut s. The hel pf ul bacteria that pop-
ul at e our i nt est i nes of t en work under conditions si mi l ar to
those experi enced by yeast at the bottom of ferment at i on vats.
Deprived of oxygen, these bact eri a produce mi nut e amount s
of ethanol as a resul t of anaerobi c metabolism. Apparently,
nat ur al selection fav ored creat ures t hat could get rid of these
t i ny quant i t i es of et hanol . Thus creat ures making enzymes ca-
pable of destroying cthanol were more likely to survive than
cr eat ur es t hat f el t t he f ul l ef f ect s of i nt er nal l y produced alco-
hol. Not only does t hi s expl ai n why humans have the abi l i t y
to sober up af t er drinking, but it expl ai ns why our enz ymatic
defenses are so easily overwhelmed: t hey were designed to han-
dle only the mi nut e amount s of et hanol secreted by bacteria
in our i nt est i nesnot the comparatively massive quant i t i es
cont ai ned in even a single shot of scotch.
The biochemical prot ect i on of alcohol-destroying enzymes is
not conf er r ed equal l y among i ndi vi dual s, however. Alcohol de-
hydrogenasel i ke all enz ymesis manuf act ur ed according to
bl uepr i nt s stored in DNA. Since everyone' s DNA is uni que,
i ndi v i dual s v ar y, sometimes st ri ki ngl y, in the ef f i ci ency and ac-
t i v i t y of t hei r alcohol dehydrogenase. One source of v ari at i on
between people is a mat t er of sex. For reasons t hat remain
unclear, alcohol dehydrogenase is less ef f i ci ent in femal e stom-
achs t han it is in male st omachs. In a recent study of this
phenomenon, the enz yme act i v i t y of men was 70 to 80 percent
great er t han t hat of women (Frezza et al. 1990). This may
account for the fact t hat , in general , women become i nt oxi -
cated sooner in response to the same dose of alcohol than men
do. It is also one of the reasons t hat the def i ni t i on of "moderate
dri nki ng" is defi ned di f f er ent l y for men and women: two st an-
Down the Hatch I 31
dard dri nks a day for men as opposed to only a single st andar d
drink a day for women (Gordis et al. 1995).
Prior to the study j us t mentioned, the di f f er ent rates of in-
toxication between the sexes was thought to occur because
women, on average, have a smaller volume of blood t han men.
An identical dose of alcohol, it was thought, would be more
concentrated in women' s blood than men' s. But Frezza and
his colleagues f ound t hat the di fference between men and
women in this par t i cul ar area has much more to do with what
happens in the stomach than in the blood. The subjects in
this study were given ethanol either orally or i nt rav enousl y.
With oral admi ni st rat i on, the ethanol was exposed to the al-
cohol dehydrogenase in the subjects' stomachs. Under these
conditions, the researchers observed si gni f i cant l y higher blood
levels of alcohol in the women compared with the men. But
when the ethanol was given i nt rav enousl y, no si gni fi cant di f-
ferences were f ound. The di f f er ences in total blood volume
between the sexes, in other words, had no detectable ef f ect s
on blood alcohol levels; exposure to stomach enzymes defi-
nitely did.
Interestingly, this heightened ethanol sensitivity in women
appears to apply only to young women. Another study, done by
German researchers, showed t hat the si t uat i on j us t described
reverses in men and women over age fifty (Seitz et al. 1990).
Alcohol dehydrogenase act i v i t y in men decreases significantly
with age, to the point where the act i v i t y drops below t hat
f ound in women of the same age. This means that men become
increasingly sensitive to ethanol as they age, ul t i mat el y ren-
deri ng them more sensi t i ve t han women.
As we' ll sec in a moment, sex isn' t the only genetic fact or
involved in alcohol sensi t i v i t y. But since you' re still in the
stomach l i ni ng watching alcohol dehydrogenase go to work
32 / Buzz
agai nst ethanol, it' s a good t i me to ment i on some i mpor t ant
nongenetic reasons t hat people mi ght v ar y in t hei r response to
a given dose of alcohol.
First of all, as many people know, food can a f f ec t the r at e
at which alcohol enters t he bl oodst ream. This ef f ect has less
to do with the type of food eaten than with the amount eaten.
When a sizable meal is consumed, the exi t valve of the stom-
acha muscul ar gate called the pylorie sphi nct ercl oses so
t hat the stomach can get to work di ges t i ng the food. This t raps
alcohol in the stomach, which not only prev ent s it from being
rapi dl y absorbed in the smal l i nt es t i ne but also increases the
chances thai: it will be destroyed by the alcohol dehydrogenase
f ound in the stomach l i ni ng. When the stomach is empt y or
cont ai ns only a small amount of food, however, the pylorie
sphi nct er relaxes and allows alcohol to pass i nt o t he small in-
t est i ne, which absorbs alcohol much more qui ckl y t han the
stomach does.
A less commonly appr eci at ed vari abl e in an i ndi v i dual ' s re-
sponse to alcohol involves aspi ri n. For reasons t hat remai n
unexplained, aspi r i n disables alcohol dehydrogenase (Risto et
al. 1990). In one study, the average blood alcohol levels of
subj ect s who consumed alcohol an hour a f t e r ingesting two
Maxi mum Bayer aspi r i n t abl et s were 26 percent higher t han
subject s who consumed ethanol without first t aki ng aspi ri n.
Clearly t hi s represent s a si gni f i cant i ncrease, part i cul arl y since
many people assume t hat t hey can have one drink an hour and
remai n sober. Aspirin changes t hi s equat i onespeci al l y among
women, for t he reasons j us t ment i oned. Other t hi ngs being
equal , t aki ng aspi r i n will resul t i n hi gher and l onger-l ast i ng
blood alcohol levels. This i mpai r ment of alcohol dehydrogenase
in the stomach can also be caused by certain drugs, such as
t he ant i -ul cer drugs ei mct i di ne and r amt i cl i ne.
Down the Hatch I 33
Central Detox
Bored now with the speetacle of alcohol dehydrogenasc ripping
into ethanol molecules in the stomach l i ni ng, you ext ri cat e
yoursel f and plop down i nt o the flow of j ui ces leaving the stom-
ach. Soon you' re through the pylorie sphincter and are sluicing
i nt o the upper reaches of the small intestine. As the wall of
the small intestine comes into view, you notice that it looks
f ur r y. This "fur" expl ai ns why alcohol is so rapidly absorbed
here, as opposed to its rel at i vel y slow absorpt i on in the stom-
ach. The walls of the small intestine are covered with millions
of mi nut e projections called villi. These st r uct ur es give the
small intestine a surface area of more than 200 square meters
roughly the size of a t enni s court . This enormous s ur f ace area
makes the small i nt est i ne an ideal place for the absorption of
small molecules like wat er , glucose, ami no acidsand alcohol.
Wi t hi n seconds, you and the load of ethanol pass through
the villi and are mixed with blood surging toward the l i v er. All
bloocl from the digestive organs, i ncl udi ng the stomach, is
shunt ed to the l i v er , which fi l t ers toxins, bact eri a, and other
pot ent i al l y ha r mf ul substances before they get into general cir-
cul at i on. Other t han the brain i t sel f, the liver is the organ most
of t en associated wi t h alcohol. And with good reason. Although
alcohol a f f ec t s every body system and every organ to some ex-
tent, the liver bears the br unt of the assault by v i r t ue of its
role as the body's detox center. Although the liver is amaz i ngl y
resilient (remove half of it and it can regenerat e f u l l y ) , it is not
i nvul nerabl e. Long-term exposure to alcohol can cause a num-
ber of cri ppl i ng diseases, i ncl udi ng ci rrhosi sa permanent
scarring caused by the death of liver cells.
As you ent er the liver, you see ethanol molecules d i f f u s i n g
qui ckl y out of the bloocl and into the surroundi ng liver cells.
34 / Buzz
You tag along behi nd one group of ethanol molecules. Dead
ahead, you notice a f a mi l i a r molecule: alcohol dehydrogenase.
This one is slightly di f f er ent from the ones you saw in the
stomach. Amaz ingly, the liver deploys not one, but seventeen
di st i nct v ar i et i es of alcohol dehydrogenase in its ef f or t to de-
f end the body against alcohol. Each v ar i ant has characteristics
t hat d i f f e r subt l y from those of the others, allowing the liver
to det oxi f y ethanol at a broad range of concent rat i ons and am-
bient conditions. Working ful l -t i l t , these enzymes can inter-
cept and disable all the roughl y 200 qui nt i l l i on ethanol
molecules in a hal f - ounce of pure ethanol in about an hour.
This f act is the basis for the one-dri nk-an-hour rule of t humb
for remaining sober.
As we've seen, however, t hi s rule must be applied car ef ul l y.
It' s most accurat e for young, heal t hy males who slowly con-
sume a modest dri nk over t he course of an hour, who are t aki ng
no other drugs (such as aspi ri n) t hat i nt er f er e with the action
of alcohol dehydrogenase, and who arc not dri nki ng on an
empt y stomach. Changi ng any of these variables means that
more time must be allowed between dri nks to ensure sobriety.
But even t hi s stringent i nt er pr et at i on of the one-drink-an-
hour rule is i ns uf f i ci ent l y narrow. It t ur ns out t hat one's race,
as it relates to the second step in the liver' s detox process, can
also af f ect one's sensi t i vi t y to ethanol.
Remember t hat acetaldehyde, the product of the first st ep
of alcohol metabolism, is rel at i v el y toxic due to its propensi t y
to bind with proteins. As j us t ment i oned, in most people, this
toxic acetaldehyde is rapidly converted into harmless acetic
acid by an enz yme called aldehyde dehydrogenase. But because
of a subt l e change in the st r uct ur e of aldehyde dehydrogenase,
t hi s second detox step is blocked in some people. Roughly half
of all Asians produce an inactive form of aldehyde dehydro-
genase (Chen and Yeh 1989). The mut at i on in the gene used
Down the Hatch I 35
to make the enzyme is ast oundi ngl y minor, given t hat aldehyde
dehydrogenase is const ruct ed from a string of 374 individual
ami no acids. One of those 374 amino acids is altered in the
mut ant , inactive form (Takeshita et al. 1993). That single
change alters the physical shape of the enzyme enough to pre-
vent the bi ndi ng of acetaldehyde, t hus rendering the enzyme
impotent. When people with the mut ant enzyme drink alco-
hol, t hei r blood levels of acetaldehyde skyrocket, producing a
symptom dubbed the "alcohol flush reaction." The dri nker' s
face fl ushes bright red, and he or she experiences heart palpi-
tations, dizziness, and nausea. All in all, it is reported to be a
very uncomfortable experience.
In a curi ous twist of racial genetics, the problems caused by
this inability to destroy acetaldehyde are exacerbated by the f act
t hat many Asians also possess an unusual l y active form of al-
cohol dehydrogenasethe enzyme t hat produces acetaldehyde
from ethanol. They are t hus dealt a genetic double-whammy:
they produce more toxic acetaldehyde than normal, but t hei r
ability to dispose of that acetaldehyde is severely limited.
Recent work has identified two subgroups of Asians with
deficient enz ymatic machinery. Those with nearly complete in-
activation of aldehyde dehydrogenase (and thus those with the
most pronounced reactions) have been termed "fast flushers,"
while those with only a moderately disabled enz ymat i c machin-
ery have a less severe reaction and are called "slow flushers."
As unpleasant as it may be to experience, the cloud of the
flushing response may have a silver lining. A study of 1,300
Japanese alcoholics found that none were f ast flushers and only
8 percent were slow flushers (Higuci et al. 1994). The aut hors
of the study conclude t hat the flushing response probably de-
ters individuals from drinkingor from drinking enough to
trigger alcoholism. The authors caution, however, that both
general alcohol intake and alcoholism itself have been rising
36 / Buzz
st eadi l y in Japan in the past two decades. They suggest t hat
env i r onment al fact ors sueh as increased stress and easier access
to alcohol owing to r i si ng s t andar ds of living may be counter-
bal anci ng t he genetic protection against alcoholism conferred
by the flushing response, par t i cul ar l y among t hose with the less
severe response.
As we will see in l at er chapt ers, genetically det ermi ned var-
i at i ons in alcohol metabolism are only one way that biology
i nf l uences response to alcohol. But as the st udy j us t cited in-
dicates, env i ronment al i nf l uences can powerful l y af f ect the way
genet i c pr edi sposi t i ons arc expressed i n human behavior.
On the Loose
Af t er passing t hrough the liver, you exit i nt o a nearby blood
v essel . As you look ar ound, you see t hat many et hanol mole-
cules hav e escaped dest r uct i on. This isn' t sur pr i si ng, since the
shot of scotch was admi ni st er ed in a single gul p on a relatively
empty st omachexact l y the t ype of onsl aught for which na-
ture did not prepare us.
You arc now headed, along wi t h t he load of ethanol, up to
t he heart and i nt o wi despread ci r cul at i on t hrough t he body.
And t hat means the end of our i magi nary j our ney. Unl i ke the
rel at i v el y const rai ned pat h f r om shot glass to liver, the paths
t aken by et hanol molecules once t hey leave the heart arc so
di v erse t hat a single v ant age point i sn' t very us e f ul . But our
j our ney has served admi r abl y t o i nt r oduce some speci fi c actions
of alcohol as well as some general operat i ng principles of the
body. We've "wi t nessed" the body' s al cohol -dest ruct i on system
in act i ont he enz ymat i c tag-team of alcohol dehydrogenase
and al dehyde dehydrogenaseand we have seen t hat this sys-
tem v ari es wi del y among i ndi v i dual s . We've also seen that gen-
der, race, age, food i nt ake, and the ingestion of drugs such as
Down the Hatch I 37
aspi ri n can af f ect a person' s response to alcohol. You' re now in
a position to tailor the one-clrink-an-hour rule of t humb for
your own use.
Our microscopic vantage point has also allowed us to ex-
ami ne some of the body's basic molecular machi nery. For in-
stance, we met some ion channelsa class of molecules we' ll
meet again shortly because they are key players in the actions
of alcohol and caf f ei ne in the br ai n.
Finally, we' ve been reminded repeat edl y that f r om nat ur e' s
point of view, anyway, alcohol is a poison to be eliminated as
quickly as possible. Which raises an obvious question: Why do
humans have such a powerful urge to consume this poison?
The answer lies in what happens when what' s left of the shot
of scotch finally reaches the brai n.
Pharmacy in a Bottle
Alcohol has t r adi t i onal l y been called a depressant. The desig-
nat i on was made because at high doses, alcohol slows down
the cent ral nerv ous system. The classic symptoms of dr unk-
ennesssl urred speech, di scoordi nat i on, di mi ni shed cognitive
abi l i t yari se from a depression of f unct i on in v ari ous part s of
the br ai n. And alcohol is l et hal at very high doses because it
depresses nerve f unct i oni ng in the brai n stem to the point t hat
br eat hi ng stops. But, as Aristotle observed more t han two thou-
sand years ago, alcohol does more t han j us t "stupefy" those
Your Brain on Alcohol I 39
who drink it. It can also "drive to a frenz y. " Intoxication can
evoke boisterousness, t al kat i veness, aggression, ribaldry, and
other behaviors more typical of a st i mul ant than a depressant .
The standard explanation for these ef f ect s is that alcohol de-
presses the "higher" cognitive abilities, such as the ability to
control emotions, t hus allowing our more unr ul y, carnal sides
to emerge.
Although it contains a grain of t r ut h, this theory rather rad-
ically misses the mark. For one thing, it begs the question of
how alcohol depresses brai n f unct i on in the first place. But,
more i mport ant , current research indicates that the central
premise of the theory is wrong: alcohol is not simply a depres-
sant that produces only apparent stimulation. In reality, alcohol
directly stimulates the brain and exerts a host of more com-
plicated ef f ect s as well.
It ' s true t hat , like et her, alcoholespecially at moderate to
high dosescan act as a general anesthetic, depressing a broad
range of central nervous system funct i ons. But alcohol also
mimics the action of the drugs cocaine, amphet ami ne, Valium,
and opium (Charness et al. 1989; Koob and Bloom 1988;
Weight et al. 1993). Like cocaine and amphet ami ne, alcohol
di rect l y stimulates cert ai n brai n cells. At low doses, it increases
electrical act i v i t y in the same brain systems af f ect ed by these
classic st i mul ant s and can lead to feelings of pleasure and eu-
phori afeel i ngs t hat may underlie much of alcohol's addictive
potential. Alcohol also works on exactly the same brain ci rcui t s
targeted by Valium; the calming, anxi et y-easi ng ef f ect s of al-
cohol closely resemble those exerted by t hi s f amous t ranqui l -
izer. And alcohol also resembles opium because it can release
our internal stores of the morphinelike compounds called en-
dorphins, t hus t appi ng into the brain' s core pleasure ci rcui t s.
In short, current research reveals alcohol as a far more complex
and interesting drug than it was thought to be. It is a regular
40 / Buzz
pharmacy in a bottle: a st i mul at i ng/ depressi ng/ mood-al t eri ng
drug that leaves pract i cal l y no circuit or system of the brain
unt ouched. This broad scope, in f act , sets alcohol apar t f r om
many other drugs. Substances such as cocaine and LSD work
like pharmacological scalpels, al t eri ng the funct i oni ng of only
one or a ha ndf ul of brain circuits. Alcohol is more like a phar-
macological hand grenade. I t af f ect s pract i cal l y ev eryt hi ng
around it.
Alcohol' s lack of speci fi ci t y makes it a somewhat maddeni ng
quar r y for research sci ent i st s. It increases the firing of some
nerve cells, or neurons, while decreasing the firing of others. It
st i mul at es some regions of the br ai n while depressing others.
And the ef f ect s it exert s can change with time and dose. Given
t hi s array of conf usi ng variables, it is remarkable t hat anyt hi ng
is known about alcohol at al l . But , in f act , an enormous amount
has been l earned, par t i cul ar l y in the past ten year s. The current
picture of what happens in your brai n when you dri nk alcohol
is oft en st ri ki ngl y at odds wi t h previ ous notions about how
alcohol works.
Pruni ng
The simplest idea of how alcohol af f ect s the brain is so en-
trenched in popular lore t hat it' s the crux of a common joke
about drinking: I' m not killing brain cells, I' m j us t pruni ng to
allow for new growth. The idea seems to be that alcohol mows
clown neurons like so much Li st cri ne, killing them "on con-
tact." Given what we've learned t hus far about alcohol, this
doesn' t seem t erri bl y far-fet ched. Af t er all, alcohol is a haz ard-
ous solvent excreted by yeast , and nat ur e finds it suf f i ci ent l y
poisonous to equi p us with an el aborat e (if rel at i vel y i nef f ec-
tive) enz ymat i c defense system agai nst it. In addi t i on, very high
concent rat i ons of alcohol can indeed kill cells, which is why it
Your Brain on Alcohol I 41
is used as a di si nfect ant . But these lethal levels are never
reached in the brain i t sel f. Legal intoxication is reached when
the concentration of alcohol in the blood reaches a mere . 1
percent. That' s a far cry indeed from the nearly 100 percent
solutions used for sterilizing. Thus although alcohol does many
things to the brain, one thing it clearly doesn't do is wipe out
neurons indiscriminately.
Recently, Grethe Jensen and colleagues proved this fact by
careful l y counting neurons in matched samples of alcoholics
and nonalcoholics (Jensen and Pakkcnberg 1993). The samples
were taken f r om people who had died of causes unrel at ed to
dri nki ng. When the two groups were compared, no significant
di f f er ences in either the overall number or the density of neu-
rons were found between the two groups.
As a statement about alcohol's action in the brain, then, the
j oke about pruni ng is wrong. But in one rat her roundabout way,
the joke cont ai ns an element of t r ut h. It implies that the brain
is so crammed with neurons that you can a f f or d to kill off
t housands with every dri nk wi t hout serious risk of dementia.
This isn' t far from the t rut h. The brain is a phenomenally com-
plex piece of biological machi nery. This is due in part to the
sheer number of part s it contains: approxi mat el y 100 billion
neurons and tens of billions of additional support cells called
glia. One way to appreciate these numbers is to sneak up to
them. It has been est i mat ed that over the course of your l i f e
you' ll lose roughly 7 percent of your brai n' s neurons from nor-
mal wear and t ear (Dowling 1992). That 7 percent, t ransl at ed
i nt o actual neurons lost, corresponds to an average daily loss
of 200,000 brain cells. In other words, you are so fabul ousl y
endowed with neurons that you can a f f or d to discard roughly
a quart er-mi l l i on of them each clay wi t hout losing your mi nd
literally.
But as impressive as such numbers are, they aren' t the real
42 / Buzz
reason the brai n is so complex. Af t er all, the liver is composed
of bi l l i ons upon bi l l i ons of cells too, but it doesn' t war r ant the
same superl at i v es. The br ai n' s compl exi t y deri v es f r om the way
those 100 billion par t s are wired up. Neurons communi cat e
wi t h one anot her by way of treelike ext ensi ons called dendrites
and axons. Generally speaking, a neuron "listens" to i ncomi ng
signals with its dendr i t es and "talks" to other neurons by send-
ing signals out al ong its axon. A single neur on can communi -
cate wi t h as many as 50,000 other nerve cells in this way. And
each of t hose connect i onscal l ed synapsesis a v i t al l y im-
por t ant l i nk in the br ai n' s i nformat i on-processi ng abilities. In
fact synapses, not whole neurons, are t hought to be the f un-
damental uni t of i nf or mat i on storage and mani pul at i on in the
br ai n.
A bet t er est i mat e, t her ef or e, of the br ai n' s t rue size can be
f ound by mul t i pl yi ng 100 billion neurons by 1,000a rather
conservative average of the number of dendr i t i c connections
that each neur on makes with anot her . The r esul t : 100 t ri l l i on
synapses. 100 trillion f unct i onal uni t s. As one neurosci cnt i st
qui pped, "100 t r i l l i on synapses, hell, you can do anyt hi ng with
t hat . That' s more than enough to cont ai n a soul" (Johnson
1991).
But t hat ' s not al l . The brain is not a pri nt ed circuit board.
Connect i ons between neurons change over time. New synapses
form; old ones di sappear. The br ai n' s "wiring" is scul pt ed by
experi ence. I nf or mat i on st reami ng in through our eyes, ears,
and other sense organs can be captured because the connec-
t i ons between neur ons can change in a split second, formi ng
new circuits. Thus the t r ue measure of the brai n' s compl exi t y
isn' t j us t the raw number of neurons or dendri t i c connections
in the brain at any given point in time, but the total possible
number of connectionsthe number of ways neurons can be
l i nked together into discrete pat t erns. This number cannot be
Your Brain on Alcohol I 43
reliably estimated, though it is widely viewed as far higher than
10
78
(that' s a 10 with seventy-eight zeros af t er i t ) , which is a
rough guess at the number of atoms in the ent i re uni v erse
(Hooper and Teresi 1987).
That' s why you don' t need to fret over your average daily
loss of 200,000 neuronsi t ' s virtually insignificant in the great
scheme of things. But if alcohol doesn't act ual l y kill neurons
if we've got so many t hat it probably wouldn' t mat t er much if
it did kill neuronst hen what, exactly, is alcohol doing to all
those cells? And why do chronic drinkers seem to show clear
signs of cognitive dys f unct i on? As with the science of f er men-
tation, it took a surpri si ngl y long time for anybody to find an-
swers to such questions.
Olive Oil Clues
In the late 1890s, two German scientists, E. Overtoil and
H. Meyer, were i nv est i gat i ng the ways that di f f er ent alcohols
dissolve in olive oil. As we've seen, ethanol is a small type of
alcohol, with a backbone of only two carbon atoms. Methanol
is smaller still, with but a single carbon. But larger alcohols
have three, f our , or more carbons. Overtoil and Meyer discov-
ered that the length of an alcohol's carbon chain is related to
its ability to dissolve in olive oil and other f at s. Simply put , the
longer the chain, the easier it is for an alcohol to mix with f at .
The interesting thing about this observation is that carbon-
chain length is also related to an alcohol's anesthetiz ing and
intoxicating powers. The longer the carbon chain, the more
potent the alcohol as an intoxicant. This led to an obvious
hypothesis: perhaps alcohol's power to intoxicate is related to
its ability as a solventin part i cul ar, its ability to dissolve the
f at t y walls of cell membranes. Unf or t unat el y for the Germans,
the i nst rument at i on of the day wasn' t up to testing this hy-
44 / Buzz
pothcsis. I' hey coul dn' t even see cell membranes, much less do
car ef ul experi ment s on t hem. And so for decades, t hei r work
with olive oil stood only as a t ant al i z i ng clue about how alcohol
worked.
It wasn' t unt i l the 1960s and 1970s t hat the tools of neu-
roscicncc became subtle enough to pick up where Overtoil and
Meyer l ef t of f . A wide range of exper i ment s in t hose decades
showed that, indeed, et hanol and ot her alcohols had a disor-
dering ef f ect on cell membr anesand the longer the alcohol
molecule, the more di sorder. Specifically, t he alcohols appeared
to fluidize membranes, to make them looser and easier to dis-
t ur b.
As we saw dur i ng our microscopic t our of the tongue, cell
membr anes come st udded with a wide range of prot ei n mole-
cules, such as the ion channel s you watched openi ng and clos-
i ng. Such prot ei ns l i t er al l y float in the membrane like so many
icebergs, and so the fluidi/ ing ef f ect of alcohols coul dat least
t heor et i cal l ycause a wide range of di f f i cul t i es. For i nst ance,
the mov ement of nut r i ent s into a cell and waste product s out
of a cell might be di sr upt ed, l eadi ng to i mpai red nerve-cell
f unct i on. Unt i l as recently as the 1980s, one or anot her version
of t hi s so-called lipid t heory of alcohol' s action held sway. In-
toxication, it was st rongl y suggested, r esul t ed f r om the disor-
ganiz ing ef f ec t of alcohol on the membr anes of nerve cells,
leading to a depression in t hei r f unct i oni ng.
Then problems began to crop up. For one thing, the con-
cent r at i ons of alcohol used in the original exper i ment s were far
higher t han those causi ng i nt oxi cat i on in humans . They were,
in fact , l et hal doses. Even more damni ng, the fluidization ob-
served wi t h r eal i st i c closes of alcohol was mi nor and could be
dupl i cat ed si mpl y by r ai s i ng the t emper at ur e of the cells a few
degrees above normal body t emper at ur e (Treistman et al.
1987). Since people wi t h mild fev ers do not become i nt oxi -
Your Brain on Alcohol I 45
cated, alcohol was clearly doing more in the brai n than simply
fluidi/ ing neuronal membranes.
In light of these di f f i cul t i es, at t ent i on has now shi ft ed away
from the membrane and toward the proteins embedded in the
membrane. And the part i cul ar proteins now under scrut i ny are
the ion channels we met earlier. We've seen that by flexing or
twisting (the exact motions are uncl ear ), these large molecules
can open or close a hole in the membrane. When open, the
channels allow electrically charged atoms (ions) to surge into
or out of a cell. This al t ers the cell's electrical properties.
Ion channels arc found in the membranes of practically all
cells, but they are part i cul arl y i mport ant for neurons. The t r af -
fic of ions into and out of neurons underlies their capacity to
generate and t ransmi t electrical signals.
A growing body of evidence shows that alcohol molecules
directly af f ect the ability of ion channels to open or close. This
basic actionnot a general fluidizing ef f ect on membranes
is now thought to be responsible for the range of behavioral
phenomena lumped under the label "intoxication" (Weight
1992).
By revealing how alcohol i nt eract s with ion channels, current
research is shedding light on many ancient questions regarding
alcohol's mode of act i onnot least of which is Aristotle' s
query about why alcohol seems to be both a st i mul ant and a
depressant.
Alcohol as Ether
Alcohol's ability to depress brain f unct i on is one of its most
obvious and pot ent i al l y haz ardous at t ri but es. Recent evidence
from a number of laboratories shows that one way alcohol
achieves this depression is by i nt er f er i ng with a t ype of ion
channel critical for the firing of neurons (Lovinger and Peoples
46 / Buzz
1993). Before proceeding, however, we need to consider br i ef l y
t he widely used phrase "firing neuron. " Since such firing lies
at the root of all the many ef f ect s of alcoholand caf f ei ne,
too, for that mat t er under s t andi ng the basics of t hi s business
rs r at her i mpor t ant .
Neurons can "fire" because they generat e a relatively large
el ect ri cal charge across their membr anes. In a sense, neurons
are like microscopic bat t eri es gradual l y st ori ng charge, and then
releasing it when fired. The electrical charge used by neurons
is carried by ionsthose electrically charged atoms mentioned
previously. When positive and negative ions are mixed to-
gether, as t hey are in most par t s of the body, t hei r charges
cancel out and the resul t is an electrically neut ral solution. But
i f positive and negative ions are separat ed and concentrated, a
charge di f f er ence is developeda di f f er ence measured in volts.
Neurons achieve a separation of charge by forci bl y moving
i ons on one side or t he other of the f a t t y cell membrane, which
is an excellent electrical i nsul at or. Special protein molecules
called ion pumps f er r y positive ions out of cells, which leaves
the out si de of the cell positive and the i nsi de negative.
The process of firing generally begins with incoming signals
from ot her neur ons . These "signals" are actually t i ny surges of
positive or negat i v e ions ent eri ng the dendri t es t hr ough ion
channel s. As a resul t , the electrical charge inside t he neuron is
in const ant flux, moving above and below the average main-
t ai ned by the ion pumps. In a very real sense, each neur on is
per f or mi ng a cal cul at i on: it is adding up the signals coming
i nt o it via the dendri t es. If negative ions predomi nat e, not hi ng
happens. The cell is already negat i v e and so adding more neg-
ative ions j us t pushes the cell f ar t her in an electrically negative
direction. But if large quant i t i es of positive ions ent er the neu-
ron, the electrical charge produced by all those ion pumps is
par t i al l y neut r al i z ed. If the cell's overall charge is neut ral i z ed
Your Brain on Alcohol I 47
below a cert ai n critical poi nt , a trigger is pulled causing a spe-
cial class of ion channels to spring open near the base of the
axont he mai n fiber carryi ng messages away f r om the cell.
These channels let in a flood of even more positive ions, which
obliterates the electrical charge at t hat spot.
The sudden collapse of electrical potential around the base
of the axon is "felt," in t ur n, by adj acent ion channels a bit
f ar t her down the axon. These channels now open. The charge
collapses at this new point, triggering yet more channels to
open f ar t her down the axon. The process continues, like the
flame of a firecracker f us e. This t ravel i ng wave of altered elec-
trical potential is called an action potential, more commonly
known as a nerve impulse. Action potentials are the st uff that
minds are made of. Af f ect the way they are generated, t rans-
mitted, or received by other neurons and you a f f ec t the mind
i t s el f .
Action pot ent i al s zip down axons at about 225 miles per
hour . When they reach the end of the axon, they don' t auto-
matically cause the next neuron in line to fire. Such an ar-
rangement wouldn' t be terribly ef f ect i v e, since a single firing
neuron would quickly ignite a crippling chain-reaction of firing
throughout the br ai n. To avoid this problem, all of the brai n' s
billions of neur ons are separated f r om each other by tiny, in-
sulating gaps called synapses (Figure 3). To cross t hi s gap, an
action pot ent i al must be converted f r om an electrical signal to
a chemical signal. It works like this:
When an action potential reaches the tip of an axon, it trig-
gers the release of one or another kind of signaling molecule
into the synapse. These signaling molecules are called neuro-
t ransmi t t ers because they t ransmi t messages between neurons.
Neurot ransmi t t er molecules are contained in t i ny bags called
vesicles inside the axon t i p. When an action potential arrives,
the bags rapi dl y f use with the cell membrane, dumpi ng their
48 / Buz
Figure 3. Neuronal synapse. (Ann Bliss P ilcher)
load of neur ot r ans mi t t er i nt o the synapse. In milliseconds, neu-
r ot r ansmi t t er molecules dr i f t across the synapse and dock into
specially designed receiving molecules on the "downstream"
neur on. This docking is a mat t er of molecular geomet ry: the
bumps and knobs of a neur ot r ans mi t t er fit into corresponding
di mpl es and holes on the sur f ace of a receiving molecule called
a recept or. Receptors come in dozens of v ari et i es, each specially
designed to accommodate one of the dozens of neur ot r ans-
rmtters used by the br ai n.
When the right neur ot r ans mi t t er docks with the appropri at e
receptor, the physical st r uct ur e of the recept or molecule
changes. Sometimes receptors are ion channels, and the dock-
ing t ri ps the channel opeir. Other times the receptor acts as a
si gnal i ng station, t ri ggeri ng a chemi cal chai n reaction inside
the cell t hat sends a message to nearby ion channel s to either
open or close. Regardless of whether it' s direct or i ndi rect , how-
Your Brain on Alcohol I 49
ever, the end resul t of neur ot r ans r ni t t er s crossing a synapse is
usually the flow of ions i nt o (or out of) the receiving neuron.
If the ion channel l et s posi t i v e ions into the recei vi ng neuron,
the neuron is pushed toward firing. But if the receptor lets in
negative ions, the downstream neuron is made more resistant
to firing. Anything that i nt er f er es with these receptors i nf l u-
ences the messages being sent from neuron to neuron in the
brai n.
Now it' s t i me for a dr i nk (fi gurat i v el y speaking, of course).
Short Ci rcui ts and Blackouts
Hav i ng seen some of the mechanical underpinnings of brain
f unct i on, we' re able to appreci at e how molecules of alcohol
might pl ausi bl y a f f ec t t hi s f unct i oni ng. Let' s st art by looking
at one way alcohol can slow clown brai n act i v i t y.
One of the maj or neur ot r ansr ni t t er s used to send "fire" mes-
sages from one neuron to another is a molecule called gluta-
mate. When gl ut amat e is released i nt o a synapse, it docks at
a receptor that lets positive ions rush in. Since this makes it
more likely t hat the recei vi ng cell will fire, glutamate is called
an excitatory neur ot r ansmi t t er .
When you take a dri nk, alcohol molecules t hat escape de-
struction in the liver arc qui ckl y pumped up to the brai n, where
they i nf i l t r at e synapses everywhere. There they can bind to glu-
tamate receptors. Nobody knows precisely where on the recep-
tor alcohol binds, but it somehow warps the st r uct ur e of the
receptor j u s t enough to i nt er f er e with its ability to open nor-
mally, t hus mut i ng gl ut amat e' s normal "fire" message. Alco-
hol's i nhi bi t i on of glutamate receptors can be pr of ound. Af t er
consumpt i on of the equivalent of about two dri nks in the space
of an hour, gl ut amat e receptor f unct i on can be reduced by-
more than 80 percent (Weight et al. 1993).
50 / Buzz
By inhibiting the br ai n' s most common exci t at ory neuro-
t r ansmi t t er , alcohol ef f ect i v el y slows down act i v i t y in many
part s of the br ai n. If the neur ons in those areas control muscl es,
the i nhi bi t i on can lead to rel axat i on and discoordination. If
the neur ons control speech, words sl ur and become i ncreasi ngl y
i mpreci se. If the neurons control aut omat i c bodily f unct i ons,
heart rate and br eat hi ng are i mpai red. From a public heal t h
point of view, these arc among alcohol' s most dire ef f ect s. The
i nhi bi t i on of gl ut amat c receptors is the molecular f oundat i on
of such grim st at i st i cs as the annual deat h of more t han 20, 000
people in alcohol-related t r a f f i c acci dent s.
The i nhi bi t i on of gl ut amat e receptors may also disable one
of our most coveted i nt el l ect ual capaci t i es: the ability to l ear n.
Although i t ' s of t en compared to a computer, the brain more
closely resembl es a tablet of wet clay into which i mpressi ons
can be made, erased, and made again over t i me. This flexibility
enables us to l earn and remember. The cur r ent t heory of mem-
ory suggests that you remember somet hi ng when a specific con-
st el l at i on of neur ons is s t i mul at ed v i gorousl y. Whether it' s a
whi f f of ci nnamon or a cat chy song, an i ncomi ng st i mul us in-
st ant l y lights up a part i cul ar const el l at i on of neur ons. If con-
di t i ons are ri ght , the connect i ons between the neur ons in the
constellation are aut omat i cal l y st rengt hened in the process. If
this pat t er n of neurons is st i mul at ed in exactly the same way
again, the net work "lights up" more easi l y t han it did previ-
ously. The original sensat i on is t hus "stored" in these discrete
pat t erns of t uned connect i ons. The more of t en a par t i cul ar
pat t er n is st i mul at ed, the more sensi t i ve and per manent the
connect i ons between the neur ons in the pat t er n become.
The technical term for such long-lasting changes in the
strength of synaptic connections is long-term potentiation, or
LTP. (A mi rror phenomenonl ong- t er m i nhi bi t i oni s also
likely to be involved in memory f or mat i on. ) 'The di scov ery of
Your Brain on Alcohol I 51
LTP in 197? provided the first plausible meehanism to support
the theory outlined earl i er. When t hi s phenomenon was in-
vestigated closely, it was discovered that LTP is blocked when
a specific ki nd of gl ut amat e receptor is disabledthe NMDA
receptor.
Di srupt i ng NMDA receptors has serious consequences. Rats,
rabbits, and other ani mal s i nj ect ed with chemicals t hat block
NMDA-receptor channels can' t l earn new tasks, such as ne-
gotiating their way through a maze, and they are incapable of
f or mi ng new memories. Their abi l i t i es r et ur n when the ef f ect s
of the chemicals wear o f f .
The salient point here is t hat , of all the gl ut amat e-recept or
channels (there are three basic types) the NMDA receptor is
the most sensitive to alcohol (Weight et al. 1993). Experi ment s
show a 30 percent reduction in LTP at alcohol concent rat i ons
reached af t er only a single dri nk (Blitzer et al. 1990). The im-
pai rment worsens with higher alcohol concentrations, stabiliz-
ing at roughly 80 percent with a concent rat i on roughly
equi val ent to seri ous inebriation: a blood alcohol level of .2
percent about twice the legal limit for i nt oxi cat i on in most
states.
This research shows t hat alcoholeven at very low doses
di srupt s the cellular machi nery most widely believed to under -
lie our ability to f or m new memories. Since the di sr upt i on can
occur at levels below those causing more obvious i mpai r ment s
of motor f unct i on and speech, people may not appreciate the
degree to which t hei r memories are being impaired.
Int erest i ngl y, the i mpai r ment is of the ability to form new
memories, not the abi l i t y to recall stored memories. Int oxi -
cated people who were asked to recall a list of words learned
prior to i nt oxi cat i on showed no i mpai rment of t hei r recall abil-
ity (Birnbaum et al. 1978). In cont rast , when the words were
presented to people already i nt oxi cat ed, their ability to recall
52 / Buzz
the words l at er dropped si gni f i cant l y (Jones 1973). Results such
as these suggest t hat the operat i ons of memory acqui si t i on and
memory r et r i ev al are separat ed in the brai n and rely on di f f er -
ent ki nds of mol ecul ar machi ner y.
The memory i mpai r ment ' r es ul t i ng from alcohol ranges from
a bar el y detectable "cocktail-party amnesi a" to the full-blown
memory bl ackout s experi enced by alcoholics. Inhi bi t i on of
NMDA channel s is the most l i kel y cause of the moderat e im-
pai rment s, but the mol ecul ar basis for alcoholic bl ackout s has
not been det er mi ned. It may be due to the combined ef f ect s
of the i nhi bi t i on of NMDA channels and the al t er at i on of ot her
types of ion channel s that produce a massive i nhi bi t i on of
nerve-cell f i r i ng in the hi ppocampus, a portion of t he br ai n
cr i t i cal t o memory f or mat i on.
If alcohol af f ect ed only neur ons and neural networks t hat
rel y on the neur ot r ansmi t t er gl ut ar nat e, it would still be a pow-
e r f ul substance. But such ef f ec t s al one would not make for a
very popul ar dr ug. I ndeed, people dr i nk alcohol despite the fact
t hat it depresses overall br ai n f unct i on and can radically i nt er-
f er e with the ability to l ear n. Accounting for alcohol' s enor-
mous popul ar i t y and expl ai ni ng i t s myri ad ot her ef f ect s
requi res t hat we look beyond gl ut amat e to some of the br ai n' s
ot her i mpor t ant ncur ot r ans mi t t er s .
Li qui d Vali um
Anxiety is an unpl easant emotional state that di f f er s f r om re-
lated states such as f ear , aggression, and conf usi on. Not only
does anxi et y feel di f f er ent , but at a purel y neurological level, it
z' s di f f e r e nt . Evidence for t hi s comes f r om experience with a
f ami l y of drugs called ben/ odi axepi nes, of which Valium is the
most well known. At low to moderat e doses, these drugs sig-
Your Brain on Alcohol I 53
ni fi cant l y reduce anxi et y wi t hout i mpai r i ng or di sr upt i ng other
br ai n systems.
Val i um wor ks by enhancing the f unct i on of a receptor t hat
plays yin to gl ut amat e' s yang in the brai n. Inst ead of passing
on a message to "fire," t hi s receptor makes it harder for a neu-
ron to fire. The receptor in question is triggered by a neuro-
t r ansmi t t er known as gamma-aminobutyric acid, or GABA.
When GABA docks at its receptor, the associated channel
opens and lets negative ions rush in, which pushes the cell even
f a r t her from i t s t ri gger point for fi ri ng. Although such inhibi-
tion might seem count erproduct i v e, it is actually crucial. Nor-
mal br ai n f unct i on depends on both exci t at ory and inhibitory
neurot ransmi t t crs. The situation is analogous to the operation
of an automobile, which requi res both an accelerator and a
brake. Gl ut arnat e is one of the brain' s accelerators, and GABA
is one of its brakes. We've already seen that by i nt er f er i ng with
gl ut amat e channels, alcohol i nt er f er es with the accelerator,
making it harder to gain speed. Now we'll see that another way
alcohol slows the brai n is by increasing the sensitivity of the
brakes. This, in fact, is how Valium and other benzodiazepines
work. These compounds bind to GABA receptors, which alters
t hei r shape and makes them three t i mes more sensitive to
GABA molecules (Ashton 1992). Valium, in other words, makes
the brain' s nat ur al "brake" three t i mes stronger.
Alcohol, like Val i um, can reduce anxi et y (at least in the short
t er m) , and it accomplishes t hi s in exactly the same way that
Valium does: by bi ndi ng to GABA receptors and enhancing
t hei r f unct i on. As with glutamate, the exact binding site of
alcohol hasn' t yet been pinpointed. But one thing is clear: it' s
di f f er ent f r om the sites used by Val i um and other benzodiaz-
epi nes. That' s why it' s so dangerous for people on ant i -anxi et y
drugs such as Val i um to dri nk alcohol. When both alcohol and
54 / Buzz
Val i um molecules bind to the GABA receptor, they warp the
channel to a much great er degree t han does ei t her drug acting
alone, producing a correspondingly larger i nhi bi t i on of neu-
ronal fi ri ng. Ignorance or disregard of this syncrgistic behavior
can be f at al .
The fact t hat alcohol mimics the calming act i ons of Valium
is wi del y seen as part of its at t r act i on as a drug. Anxi et y is so
omni pr esent in t oday' s society that it is hardl y sur pr i si ng t hat
people t ur n to a readi l y available, legal, and rel at i vel y i nexpen-
sive anxi et y-reduci ng dr ug for r el i ef .
And yet alcohol i nt oxi cat i on involves more t han mere rel i ef
f r om anxi et y. As Ari st ot l e noted, alcohol also appears to exert
directly positive ef f ect s: stimulating, euphori c, pleasurable feel-
ings not accounted for by its abi l i t y to banish anxi et y or induce
a general sedat i on. The experience of alcoholics also indicates
t hat alcohol does more t han relieve pai nt hough that may
cert ai nl y be a component of its addi ct i v e qual i t y. Alcohol in-
duces a powerful crav i ng for a ki nd of pl easurabl e feeling not
at t r i but abl e to its ef f ec t s on gl ut amat e and GABA ion chan-
nels. U nder s t andi ng these l at t er ef f ec t s requi res a bri ef look at
the br ai n ci r cui t s of bliss.
Raiding the Pleasure Center
It is cri t i cal l y i mpor t ant for ani mal s to di scr i mi nat e between
behaviors that enhance t hei r chances of surv i v al and behaviors
t hat under cut those chances. Ani mal s possessing some kind of
i nt er nal compass to help them make these choices would
clearly have an advantage over ani mal s forced to l earn which
behav i ors are adapt i v e and which ar en' t . It is not surprising,
t her ef or e, that brai n ci r cui t r y has evolved t o per f or m this f unc-
tion.
In the broadest sense, t her e are two such systems: rewarding
Your Brain on Alcohol I 55
ci rcui t s and puni shi ng ci rcui t s. Like a biochemical carrot and
stick, these systems generate pleasurable or pai nf ul feelings
that powerful l y guide behavior. The reward ci rcui t s generate
cravings that impel an ani mal toward such things as eating,
drinking, and procreating. When one of these actions is com-
pleted successful l y, neurons in a specific part of the brain re-
lease chemicals that elicit feelings ranging f r om a calm satiety
to orgasm.
The existence of a discrete reward center in the brai n was
first demonstrated in 1954 by physiologist James Olds, who
placed very fine electrodes in the brai ns of rat s and allowed the
animals to st i mul at e cert ai n areas of their brains by pressing a
lever in their cage. He found a dramatic response when the
electrodes were placed in a region called the mesolimbic area.
Rats with electrodes in t hi s area seemed to enj oy the st i mu-
lation v ery much and worked very hard to obtain it, even if
this meant learning to negotiate a complex maze. When al-
lowed to st i mul at e themselves at will, they would sometimes
do so at the rate of over a hundred times a mi nut e for hours
on end. In f act , some ani mal s starved themselves rat her t han
give up pressi ng the lever.
The electrodes gave the rat s access to t hei r own stores of
bliss-producing neur ot r ansmi t t er s. Normally meted out f r ugal l y
and only af t er the accomplishment of some i mport ant survival-
enhanci ng task, these compounds were now available at the
press of a lever. Allowed f r ee access to their own reward centers,
many of the rats became hopeless lever addicts. Of course,
implanted electrodes aren' t the only way to stimulate the
brain' s mesolimbic reward center. Cocaine, heroi n, amphet a-
mine, nicotine, and a great many other drugs give humans a
lever for accessing their pleasure centers.
The role of the mesolimbic system in addiction and drug use
is so compelling t hat it is the focus of intense research. The
56 / Buzz
syst em is ext r emel y compl ex, and we are far from a complete
under s t a ndi ng of i t s f unc t i on. Yet, t hanks i n part t o research
on drugs such as cocaine and opi um, some of the basic neu-
r ot r ansmi t t cr systems vital to the reward center have been elu-
ci dat ed.
Alcohol v ery l i kel y a f f e c t s all the neur ot r ansmi t t er s used in
t hi s center. To date, two of these hav e received part i cul ar at-
t ent i on: clopaminc, and t he opi uml i ke endorphi ns. By altering
t hese neur ot r ans mi t t er s , alcohol is now t hought to evoke the
euphoric, hedonic sensat i ons associated with i nt oxi cat i on.
These are also the brain chemi cal s widely regarded as most
i nt i mat el y involved in alcohol' s high pot ent i al for abuse among
cer t ai n dr i nker s.
Alcohol modestly i ncreases cloparnine levels in the reward
ci r cui t s of the brai n, making it a weak cousin of cocaine and
amphet ami ne (Di Chi ara and I mper at o 1988). This release of
dopami ne is t hought to underl i e the i ni t i al l y st i mul at i ng, en-
ergi z i ng feel i ngs of t en experienced by dr i nker s.
The "high" one gets f r om alcohol is, of course, cjuite di f f e r -
ent f r om that achieved by cocaine and amphet ami ne. These
drugs are much more pot ent and arc practically surgical in t hei r
ef f ect s. They / ero in on dopami ne while leaving other neuro-
t r a ns mi t t er s unt ouched. The st i mul at i on produced by alcohol,
in cont rast , is modest to begin with and must compete with
t he s i mul t aneous depr essant ef f ect s caused by t he i nhi bi t i on
of gl ut amat e channels and the enhancement of GABA chan-
nel s.
Alcohol' s ef f ect on dopami ne l ev el s has been f ound to be
most pronounced in the hrst t went y mi nut es of exposure
( Fr i edman et al. 1980, Frye and Breese 1981). This may expl ai n
why t he earl y stages of i nt oxi cat i on feel qual i t at i v el y d i f f e r e n t
f r om l at er stages. In ani mal s, t he i ni t i al "hit" of dopamine f r om
i ngest i on of alcohol correl at es with a br i ef i ncr ease in act i v i t y,
Your Brain on Alcohol I 57
which then declines to levels below that displayed before the
alcohol was admi ni st ered.
Exactly how alcohol boosts dopami ne levels isn' t known. Al-
cohol might act di rect l y on dopaminc receptors, making them
more sensitive than normal in a manner analogous to the sen-
sitiz ation of GABA receptors. Alternatively, alcohol might act
indirectly by af f ect i ng neur ons impinging on the reward center,
rat her t han acting on the reward center i t sel f . Whatever the
mechanism, alcohol' s ability to act i v at e the dopaminc ci r cui t s
in the brain' s reward cent ers provides the first good expl anat i on
for its st i mul at i ng ef f ect s since Aristotle commented on these
ef f ect s two thousand years ago.
The other mediators of pleasure being actively investigated
are endor phi ns, the body's nat ur al pai nki l l er s. During times of
severe stress or i nj ur y, endor phi n molecules are released from
the pi t ui t ar y gland and block pai n messages ar r i v i ng from var-
ious part s of the body. Secondary to t hi s i mport ant task, en-
dorphins also trigger the release of dopamine in the brai n' s
mesolimbic reward center, which, as we' ve seen, directly elicits
pleasurable feelings. Endorphin release is t hus doubly reward-
ing: it dampens pain and produces, via dopamine, a mild
"high."
Given what we've already l earned, it probably won' t come as
a surpri se t hat alcohol has been f ound to trigger the release of
endorphi ns from the pi t ui t ar y gland (Gi anoul aki s et al. 1990).
If t hi s were alcohol's only ef f ect , dri nki ng it would produce a
subtle "high" similar to that felt by marat hon r unner s and
other athletes who come by their endorphins nat ural l y. As it
is, dri nkers experience an endorphin boost simply as one of
many elements in a very potent mixyet anot her dimension
in the subject i ve experience of intoxication.
Alcohol t hus resembles opium and its deri v at i v es morphi ne
and heroin, all of which target the endorphi n system. Alcohol
58 / Buzz
is much less pot ent than opiates, however, because it works in
an ent i rel y di f f er ent way.
Opiate molecules fit snugly into the molecular receptors de-
signed for endorphi ns. They are, essent i al l y, f ake cndorphins.
Opiate users t hus can give themselves an "endorphin rush" far
more intense than anyt hi ng possible with only their own nat-
ur al suppl y of these pl easure compounds. But et hanol mole-
cules don' t look anyt hi ng like endorphi n molecules. They aren' t
f ake endorphi ns at all. All alcohol can do is tap i nt o one's
exi st i ng store of cndorphins. Since no new endor phi ns or en-
dor phi n look-alikes arc added to the system, the opiate-like
high achievable with alcohol is l i mi t ed. As with dopami ne, the
precise mechanisms behind alcohol-induced endorphin release
aren' t yet known.
An Unfinished Picture
This chapt er has present ed a por t r ai t of alcohol's act i ons in the
brai n t hat i s st art l i ngl y di f f er ent f r om t hat commonly held. Far
from being a simple depressant , alcohol is a subtle, complicated
drug t hat exerts a wide range of pharmacological ef f ect s .
We've seen t hat by i nhi bi t i ng gl ut arnat e receptors, alcohol
induces a general sedation and si gni fi cant l y impairs the brai n' s
abi l i t y to store new memori es. By increasing the sensi t i v i t y of
GABA receptors, alcohol mi mi cs Val i um and reduces anxi et y.
Like a weak version of cocaine or amphet ami ne, alcohol boosts
dopamine levels, pr oduci ng a bri ef period of heacly st i mul at i on.
And by rel easi ng cndorphi ns, alcohol resembles opi um, giving
users a r ush of pl easure si mi l ar to t he "nat ur al high" experi-
enced af t er a vigorous wor kout .
As comprehensi v e as t hi s l i st is, however, it probably does
not tell the f ul l st ory of alcohol' s ef f ect s on the brain. The brain
uses at l east f or t y neur ot r ans mi t t er s , which act on more t han
Your Brain on Alcohol I 59
one hundred types of' receptors. Scientists are still l earni ng how
alcohol alters the way many of these t r ansmi t t er s and receptors
work, and some of these al t erat i ons may t ur n out to be j ust as
si gni fi cant as t he ones considered her e.
One i mpor t ant research subj ect is the neur ot r ansmi t t er
serotonin, the t arget of the widely used ant i depressant drug
Prozac. By boosting serot oni n levels, Prozac can alleviate de-
pression, enhance mot i v at i on, and increase self-confidence.
Prel i mi nary st udi es suggest t hat alcohol also acts on the sero-
tonin system. It has been f ound, for instance, that moderately
high doses of alcohol increase the electrical current associated
with one type of serot oni n receptor by almost 60 percent (Lov-
ingcr and Peoples 1993). This increased current is functionally
equi val ent to boosting levels of serotonin in the synapseex-
actly what Prozac docs (Weight 1994).
Another l i ne of research l i nki ng alcohol and serotonin in-
volves r at s bred for t hei r avid preference for alcohol. The brains
of these rat s have si gni f i cant l y lower serotonin levels than do
the br ai ns of r at s t hat don' t crave alcohol. One hypot hesi s: the
f or mer group may like alcohol so much because it helps com-
pensate for t hei r genet i cal l y f aul t y serot oni n machi nery. In hu-
mans, some alcoholics have lower amount s of serotonin
breakdown product s in t hei r cerebrospi nal fluid t han do non-
alcoholics, i ndi cat i ng that t hei r brai ns are ma nuf a c t ur i ng less
serot oni n t han normal .
Finally, Prozac and si mi l ar serotonin-boosting drugs have
been shown in several st udi es to modest l y reduce alcohol con-
sumpt i on among both alcoholics and nonalcoholics (Amit et
al . 1984; Lawrin et al. 1986). These observations have led to
e f f o r t s to use Prozac as a supplement to more t radi t i onal al-
coholism t r eat ment programs.
Despite such findings, alcohol is clearly di f f e r e nt f r om Pro-
z ac. For one thing, whereas Prozac of t en alleviates depression,
60 / Buzz
alcohol almost always exacerbat es the feelings of hopelessness
and i ner t i a associated with cl i ni cal depressi on. For anot her , it
generally t akes f our to six weeks for Prozac' s positive ef f ect s to
"kick in," whereas alcohol' s ef f ect s are f el t very rapi dl y. At the
moment , t her ef or e, too l i t t l e is known about t he mol ecul ar and
neurological mechani sms of serot oni n to say anyt hi ng defi ni -
tive about how alcohol is rel at ed to that syst em.
Serotonin is j us t one of the ncur ot r ans mi t t cr s under inves-
t i gat i on as ncur osci ent i st s cont i nue to expl ore how alcohol
works in the br ai n. New sites of action are sure to crop up as
this i nv est i gat i on cont i nues . But t he mol ecul ar mechani sms
we've explored in t hi s chapt er go a long way t oward expl ai ni ng
some of the age-old mysteries of drinking and intoxication.
Beyond the Brai n
We now know t hat alcohol is a much more i nt er est i ng drug
than the simple depressant it is commonly t hought to be. It
produces complicated, of t en paradoxi cal ef f ect s in the brain
t hat mix and overlap with one another to create an equal l y
complex i nt oxi cat i on. Dr i nki ng alcohol can push one's mental
st at e in practically any di rect i on, f r om a st i mul at ed, energetic
euphoria to a dark, brooding hopelessness.
But the brain is hardly the only organ af f ect ed by alcohol.
It' s j ust the one most obviously af f ect ed. The alcohol in a shot
62 / Buzz
of l i quor , a glass of wine, or a bottle of beer i nf i l t r at es every
nook and cranny of the body, and it provokes changes in other
organs that can be j us t as complex and paradoxi cal as those in
the br ai n.
Most people, of course, dri nk alcohol for its mi nd-al t eri ng
properties. But t hat ' s not the only reason people reach for a
bottle. For instance, many a homemade cold remedy cont ai ns
alcohol in one f or m or another, owing to the belief that a small
dose has rest orat i v e powers. Alcohol is also wi del y t aken as a
"nightcap" to i nduce sleep. Al cohol part i cul arl y red wi nei s
downed these days wi t h the t hought t hat modest consumpt i on
conf er s prot ect i on f r om hear t disease. And alcohol i s undoubt -
edly the world' s most widely used aphrodisiac.
But does alcohol real l y enhance the pleasures of sex? Is it
good for the hear t ? Will it help you sleep better or recover
f r om a cold more qui ckl y? As with alcohol's ef f ect s in the brai n,
the answers to these quest i ons are both more i nt er est i ng and
less clear-cut t han is commonly t hought .
Le Paradoxe Fran^ais
In J anuar y 1996, the Uni t ed States government unveiled new
gui del i nes for a heal t hy diet. Prepared by a committee from
the Depart ment of Agri cul t ure and the Depar t ment of Health
and Human Services, the gui del i nes drew at t ent i on because,
for the first time, they acknowledged the positive health ben-
efi t s of moderate alcohol consumpt i on. "Alcoholic beverages
have been used to enhance the enj oyment of meals by many
societies t hr oughout human hi st ory, " the report noted, "and
accumul at i ng evidence suggests t hat moderat e dri nki ng may
lower the ri sk of heart at t acks . " This stance cont rast s shar pl y
with the l i ne t aken in the 1990 guidelines t hat "drinking has
Sex, Snores, and Stomach Aches I 63
no net health benefit." The guidelines do not encourage people
to dri nkand the recognition of alcohol's health benefits are
careful l y hedged with war ni ngs t hat no alcohol should be con-
sumed by children, teenagers, women who are trying to con-
ceive or who are pregnant, and anyone planning to drive or
who cannot control dri nki ng to moderate levels. Still, the an-
nouncement was a dramat i c and highly public endorsement of
a sizable body of medical research into the health benefi t s of
alcohol in general, and red wine in part i cul ar.
Such studies began in earnest in the 1970s when scientists
began taking note of the so-called French paradox. In studies
at t hat time comparing the diets and rates of illness in di f f er ent
count ri es, the Frenchwho consume a lot of cheese, cream,
meat, and other hi gh-fat foodshad one of the lowest inci-
dences of maj or heart disease. Only the Japanesewhose diet
is high in rice, fish, and other low-fat foodshad less heart
disease. Since France has a high per capita consumption of
wine, researchers began to look at whet her the two f act s were
related. The resul t s from more than a decade of st udy strongly
suggest t hat , in f act , t here is a connection. It now appears t hat
daily consumption of one or two st andard alcoholic dr i nks re-
duces the risk of heart disease. The evidence also suggests that
red wine, in par t i cul ar , is good for the heart.
Does this mean that doctors are now prescribing wine for
patients at risk for heart attacks? Hardly. The issue has pro-
voked car ef ul l y worded articles in medical j ournal s about the
proper advice to give pat i ent s who ask about wine's benefits.
For instance, an editorial in the New England Journal of Med-
icine notes that "there now seems little doubt t hat alcohol
exerts a protective ef f ect against coronary heart disease"
(Friedman and Kl at sky 1993). Yet af t er reviewing the com-
plexities of the issue, the editorial stops short of a recomrnen-
64 / Buzz
el at i on that pat i ent s begin dri nki ng. "As in ot her areas of health
care," t he a ut hor s demur , "the pat i ent must , wi t h our guid-
ance, make the final decision."
Several well-designed st udi es of heart di sease have f ound
t hat people who dr i nk no alcohol have a slightly higher l i f et i me
risk of coronary heart disease t han people who consume l i ght
to moderate amount s of alcohol. But when consumpt i on rises,
risk rises as wellto levels much hi gher t han those faced by
abst ai ners. Early studies showing this risk curve were f aul t ed
for i ncl udi ng among the t eet ot al ers people who r ecent l y quit
dri nki ng. But more rigorous st udi es el i mi nat i ng such people
from t he survey and controlling for other pot ent i al l y conf ound-
ing variables such as diet and smoking came to the same con-
clusion: moderat e dr i nki ng appears to reduce the ri sk of
at heroscl erosi s (clogged ar t er i es ) and myoeardial i nf ar et s (heart
at t acks). Some st udi es calculated a ri sk r educt i on of as much
as 50 percent; a more conservative figure, derived f r om a num-
ber of similar st udi es, is roughl y a 35 percent reduction (Fried-
man and Kl at sky 1993).
Of course, it is one thing to find an associ at i on between two
variables; it is qui t e anot her to prove t hat alcohol causes a
reduct i on in heart disease. Making this leap r equi r es a satis-
f yi ng expl anat i on of how alcohol can hav e a pal l i at i v e ef f ect on
the car di ov ascul ar system. Three possible expl anat i ons have
been put f or war d, each backed up by solid research and none
likely to be the sole mechanism. As our di scussi on of the brain
revealed, alcohol af f ect s nearly ev er yt hi ng it touches. So it
would not be sur pr i si ng if it conferred its benefi t s on the hear t
by al t er i ng several things at once.
One of the earl i est theories has also been one of the most
cont ent i ons. Ini t i al st udi es showed t hat alcohol boosted levels
of hi gh- dens i t y l i poprot ei ns ( I l DLs ) i n t he blood. I I DLs are
considered "good" because t hey t r ans por t cholesterol f r om t he
Sex, Snores, and Stomach Aches I 65
blood to the liver, where it is either t ransformed or destroyed.
The more HDL, in other words, the greater your body' s abil-
ity to move cholesterol out of your arteries. In cont rast , low-
density lipoproteins (LDLs) work in somewhat the opposite
manner, t r anspor t i ng cholesterol away from the liver and out
to peripheral tissues. Although this f unct i on is j ust as critical
to health as that per f or med by HDLs, LDLs have been labeled
"bad" since at high levels they of t en dump t hei r loads of cho-
lesterol on ar t er y walls, where it can accumulate i nt o sticky,
clogging pl aques.
If alcohol raises HDL levels, as the initial research suggested,
it could pl ausi bl y explain why alcohol appears to be so "heart
healthy." But then, as so oft en happens in science, more re-
search complicated the picture. It t urns out that there are sev-
eral ki nds of HDL. At first, it was thought that only one kind,
HDL
2
, was beneficial, and alcohol apparent l y did not af f ect
HDL,; instead, it seemed to work on HDL
3
. It has t aken years
to sort the whole thing out, but researchers now believe that
not only are both HDL
2
and HDL, i mport ant in reduci ng the
risk of heart disease, but alcohol raises the levels of both. Thus
the research has come f ul l circle. Alcohol' s ef f ect on HDLs
(both ki nds ) is once again t hought to account at least in part
for its beneficial ef f ect s on the heart .
Meanwhile, other researchers were focusi ng on red wine,
rather t han on all alcoholic beverages. One of the reasons for
this at t ent i on was that people in count ri es such as Scotland,
Finland, and the United States who consume more of t hei r
alcohol in the form of beer and hard l i quor have higher mor-
tality rates from heart disease than do the French and I t al i ans,
who consume most of t hei r alcohol in the form of wine, par-
t i cul arl y red wine. The research conducted to dat e has t ur ned
up several i nt r i gui ng resul t s.
John Folts and his colleagues at the Uni v er si t y of Wisconsin
66 / Buzz
Medical School have f ound convi nci ng evidence t hat a class of
compounds collectively called phenol s dramat i cal l y reduces the
ability of blood pl at el et s to cl ump together into clots. Specif-
ically, Folts f ound t hat two kinds of phenol squercet i n and
r ut i nabol i sh or si gni f i cant l y reduce t he cl ot - f or mi ng abi l i t y
of pl at el et s in dogs. Red wines cont ai n much higher percent-
ages of phenols t han do white wines; when Folts tried the ex-
peri ment with whi t e wi ne, he f ound little ef f ect . Thus one way
t hat red wi ne, at least, may help is by reduci ng the f or mat i on
of pot ent i al l y deadl y art eri al clots and clogs.
The French paradox might also be rel at ed to recent findings
from Queen El i / abet h Hospi t al in Bi r mi ngham, Engl and. Re-
searchers t her e were i nt ri gued by findings t hat red wine, in test-
tube experi ment s, was f ound to be an ant i oxi cl ant . Oxi dat i on
is si mpl y the reaction of oxygen with some ot her compound.
Fire is an example of r api d oxi dat i on, while rust is indicative
of slow oxi dat i on. In your body, oxi dat i on of glucose is essential
for energy product i on. But sometimes oxidation is not so help-
f ul . When some compounds are oxidiz ed, they become reactive
and unst abl e. In this state, they can easily damage or di srupt
near by molecules. One molecule prone to this ki nd of dam-
aging oxidation is l ow-densi t y l i poprot ei n. Oxidized LDL reacts
with prot ei ns and ot her compounds in cells, i nt er f er i ng with
nor mal f unct i oni ng.
Which brings us back to red wine. The Queen Eli/ abeth
Hospital researchers knew t hat red wine blocked the oxi dat i on
of LDL in the test tube, but they want ed to know if it did the
same t hi ng in huma n beings. In an exper i ment with five men
and five women, t hey f ound that the levels of ant i oxi dant s in
the blood rose consi derabl y a f t e r v ol unt eer s ate a meal and
dr ank two 5-ounce glasses of Bordeaux reel wine. When the
same meal was consumed wi t hout red wi ne, ant i oxi dant levels
act ual l v fel l .
Sex, Snores, and Stomach Aches I 67
Whichif anyof these three mechanisms is the one pri-
marily responsible for the French paradox remains to be seen.
It may be that all three cont ri but e to the overall ef f ect .
With all this evidence suggesting that moderate consump-
tion of wi neand probably other f or ms of alcohol as well
confers protection against heart disease, why isn' t everyone
reaching for their fav ori t e bottle of cabernet? There are several
reasons. First of all, the est i mat ed 35 percent reduction in risk
is offset to some extent by slight increases in the risk of con-
tracting one of the many diseases associated with drinking al-
cohol, or of being involved in an alcohol-related accident. The
French, while enjoyi ng their much reduced rates of heart dis-
ease, develop liver disease at a rate that is roughly twice that
of Americans (Dolnick 1990). In addition to t axi ng the liver,
moderate dri nki ng has been associated with a slightly increased
risk of breast cancer and cancer of the bowel. And, of course,
even a single shot of l i quor consumed quickly can produce
transient blood alcohol levels high enough to reduce reaction
times and impair coordination, t hus increasing the risk of ac-
cidents.
Second, advising abstainers to begin drinking could lead to
increased alcoholism because it is not yet possible to predict
who will succumb to alcohol's addictive potential. And fi nal l y,
drinking modest amount s of wine or other types of alcohol is
hardl y the only way to reduce the risk of heart disease. Other
methods, such as losing weight, qui t t i ng smoking, and exercis-
ing, of f er even greater benefits and have fewer associated risks.
The moral of the French paradox is that if you don' t drink,
don' t start j us t to help your heartit' s not worth it. If you
drink moderately, current research suggests that you needn' t
worry t hat you' re hur t i ng your hear t i n fact , you' re probably
helping it. And if you drink more than two or three dr i nks a
day, you should probably cut back, if for no other reason than
68 / Buzz
dri nki ng this much i sn' t going to help your cardiovascular sys-
tem and will probably compromise your overall health.
And I in My Cap . ..
A lot of' people have a hard t i me getting to sleep at night
roughly 30 million people in the Uni t ed States alone, according
to one estimate (Palca 1989). Stress, emotional upheaval , de-
pr essi on, medi cat i ons, not to ment i on excess caf f ei ne can all
leave you st ar i ng at the ceiling unt i l the wee hours of the morn-
ing.
No one knows how many insomniacs t ur n to the nearest
bottle of booze in an ef f or t to get some sleep, but the mere
f act t hat a late-night nip is known as a "nightcap" suggests
t hat this is a common use of alcohol. At first glance, t hi s seems
to make sense. Many people have fel t sleepy af t er dri nki ng and,
as we' ve seen, alcohol depresses some brain ci rcui t s by en-
hanci ng t he act i v i t y of GABA recept ors. In f act , the most
widely prescribed class of sl eepi ng pills are benz odiaz epines
such as I l al ci on, Valium, Xanax, and Rest ori l , which work by
enhanci ng GABA-reccptor f unct i oni ng.
But alcohol act ual l y makes a rat her poor sleeping pill. It may,
i ndeed, nudge you into dr eaml and, but you don' t necessari l y
stay t her e and you have a very good chance of waking in the
mor ni ng feel i ng decidedly un-rest ed.
As we' ve seen, alcohol is bot h a depressant and a st i mul ant .
Among ot her t hi ngs, it boosts dopami nc and cndorphi n levels,
both of which can elicit st i mul at i ng or mildly euphori c sensa-
t i ons. These ef f ect s are part i cul arl y pronounced at relatively
low dos es j us t t he ki nd of doses t ypi cal of a "nightcap." Thus
a single shot of whi skey or a small glass of wine t aken j us t
before bedtime may have an ef f ect t hat is exact l y the reverse
of the one being sought by the dr i nker .
Sex, Snores, and Stomach Aches I 69
More si gni f i cant l y, controlled sleep st udi es have shown t hat
v ol unt eers who consume moderat e to heavy amount s of alcohol
before going to bed tend to drop off to sleep relatively qui ckl y,
but t hen wake of t en in the middle of the night and have di f-
ficulty going back to sleep (Stradling 1993). The best expla-
nat i on for t hi s is the so-called rebound ef f ect , ot herwi se known
as acute tolerance.
Tolerance is the adj us t ment of the brain to the presence of
a drug, necessitating larger and larger doses to achieve the same
ef f ect s as the original dose. Tolerance has been observed in
humans with every drug of abuse, including alcohol and caf-
f ei ne. The tolerance t hat develops from long-term, repeated
exposure to a drug is called chronic tolerance, and it's the basis
for many of the severe problems associated with drug addic-
tions. But tolerance can also develop af t er as little as a single
dose of a dr ug (Iv ersen and Iversen 1981). In the case of al-
cohol, for instance, the brai n can adapt very rapidly, changing
within hours to counteract the resul t i ng imbalances. Such tol-
erance is relatively short-lived, but it' s enough to di st ur b sleep
in some i mport ant ways.
One way is by di srupt i ng an i mport ant phase of sleep nick-
named REM, which st ands for rapid eye movement. The
brain' s electrical activity duri ng REM sleep looks almost iden-
tical to that observed when a subject is wide awake. Heart and
breathing rates are highly variable, the eyes move rapidly under
closed lids as if the person were wat chi ng a movie, and, if
awakened, subj ect s of t en report that they were dreami ng. Peo-
ple t ypi cal l y have between f our and six peri ods of REM sleep
a night and spend about 25 percent of t hei r total sleeping time
in t hi s stage.
Alcohol consumption tends to reduce the amount of time a
person spends in REM sleep. The reasons for and i mport ance
of this finding remai n uncl ear, but REM sleep has been found
70 / Buzz
to be critical for l earni ng speci fi c kinds of new t asks (Kami et
al. 1994)possibly because duri ng REM sleep the brain "re-
plays" the day' s events in a way t hat sol i di fi es and consolidates
l earni ng and memory. This ef f ect of REM sleep seems partic-
ul arl y i mpor t ant for so-called procedural memory, which is
what we use when we learn to ride a bike or touch-type. In the
Kami st udy, subj ect s deprived of REM sleepbut not other
phases of sl eepmade no progress in l ear ni ng certain proce-
dur al t asks, whereas vol unt eers allowed REM sleep but de-
prived of non-REM sleep improved t hei r per f or mances
overnight.
It is likely that non-REM sleep plays an i mport ant role in
memory and learning as well, though in di fferent ways t han
REM sl eep. Both t ypes of sleep, however, are di sr upt ed by al-
cohol. Dur i ng the early phases of the night, alcohol reduces
REM sleep. Later on, however, the rebound ef f ect leads to
restlessness, which i nt er f er es with both REM and non-REM
sleep.
The di sr upt i v e ef f ect s of alcohol on sleep can be exacerbated
by c a f f ei ne. Caf f ei ne is broken down by the liver much more
slowly t han alcohol. It t akes about five hours for the liver to
metabolize half of a given dose of c a f f ei ne (St radl i ng 1993).
This has some i nt erest i ng i mpl i cat i ons for people who at t empt
to use alcohol as an ant i dot e to the wakef ul ness induced by
too much caf f ei ne. The sedat i ng ef f ec t s of a moderate to strong
dose of alcohol may, at first, override the st i mul at i ng ef f ect s of
the c a f f ei ne, promoting sleep. But by the middle of the ni ght ,
the alcohol will be metabolized and the rebound ef f ect will set
in. Meanwhile, the caf f ei ne will still be in circulation, which
exacerbates t he mi l d st i mul at i on resul t i ng from t he rebound
ef f ect , '['he resul t is a biochemical doubl e-whammy that can
leave you awake in the mi ddl e of the night and groggy the next
Sex, Snores, and Stomach Aches I 71
morning, which sets the stage for even more caf f ei ne dri nki ng
and a perpet uat i on of the cycle the following night.
To Your Health?
The use of alcohol as a rest orat i v e medicine is as ancient as its
use as an intoxicant. For instance, the New Testament book
of Timothy cont ai ns the suggestion to "drink no longer water
but use a little wine for thy stomach' s sake and t hi ne i nfi rmi -
ties" (1 Timothy 5: 23) .
When the art of distillation was discovered in the Middle
Ages, the potent ext ract t hat resulted was deemed both a good
medicine in i t sel f and an ideal base for the creation of other
remedies by the addi t i on of herbs and other ingredients. In
f act , the original name for alcohol was aqua vitae, Latin for
"water of l i fe. " It was regarded as a life-giving, l i f e- af f i r mi ng
l i qui d.
The idea that alcohol had medicinal qualities persisted well
i nt o the twentieth cent ury. Even during America' s periodic fits
of prohibition, the members of temperance societies were
asked only to forswear alcohol as a beverage, not as a medicine
(Tice 1992). Physicians in the eighteenth and nineteenth cen-
t uri es administered many of their medicines by dissolving them
in wine made from grapes, elderberries, blackberries, or apples.
Rye whi skey, mixed wi t h rock-sugar syrup, remained a popular
cough remedy i nt o the early t went i et h cent ury. In the South,
many believed that mint j ul ep prevent ed mal ar i a. And in the
mi d-ni net eent h century, a wide range of patent medicines with
alcohol contents averaging around 40 percent (about the same
as scotch) were bought by millions seeking cures for everything
from baldness to gout. One part i cul arl y popular brand, Lydia
Pinkham' s Vegetable Compound, was 20 percent alcohol and
72 / Buzz
carried on each bottle the slogan "Trust Lydia Pinkham, not
the doctor who doesn' t under st and your problems." The tonic
was market ed pri nci pal l y to housewives and grossed appr oxi -
mately $300,000 in one year, maki ng Pi nkhar n one of the
richest women of her day.
Surpri si ngl y, Pi nkham may have been on to something. Even
though the alcohol content of pat ent medi ci nes such as Pink-
harn' s was relatively high, the amount generally taken was small
(at least for those who real l y took the subst ances as medi ci ne),
and recent research suggests that low doses of alcohol may
confer some immunological benefi t s. Before we take a look at
these i nt r i gui ng findings, however, it' s worth poi nt i ng out the
st ark di f f er ences between acute (short -t erm) use and chroni c
use. Practically all the news from research i nt o the i mmuno-
logical impacts of chronic alcohol use is negative. Heavy drink-
ers, in addition to greatly increasing their ri sks for everything
from liver disease to high blood pressure, are also far more
susceptible to i nfect i ous diseases. Addi t i onal l y, long-term use
of alcohol depresses the i mmune syst em by i nduci ng mal nu-
t ri t i on, vitamin loss, and general i ncapaci t at i on of the liver.
These ef f ect s can leave a heav y dr i nker vulnerable to a host of
diseases ranging from cirrhosis to cancer, and serious brai n dis-
orders such as the memory-destroying Kor s akof f ' s syndrome.
There is little doubt, t her ef or e, t hat heavy dr i nki ng is bad
for one's health in general, and bad for one' s i mmune system
in particular. There is qui t e a lot of doubt , however, about the
ef f ect s of light to moderat e dr i nki ng. The evidence here is
mi xed. For i nst ance, st udi es of alcohol's effect s on nat ur al
killer cellsthe "scavenger" cells t hat dest roy v i r us- i nf ect ed
and cancerous body cel l shave produced unev en resul t s. One
st udy showed that nat ur al killer cell activity was suppressed in
mice t hat ingested high doses of alcohol over a two-week peri od
Sex, Snores, and Stomach Aches I 73
(Meadows et al. 1989). But another study, using low doses of
alcohol, showed that alcohol actually enhanced nat ural killer
cell activity (Saxena et al. 1981).
The most intriguing recent study comes from the Common
Cold Unit of Britain' s Medical Research Council (Cohen et al.
1993). This study is significant because it was unusual l y well
controlled and involved a large number of research subjects:
417. Volunteers were given in-depth physical exams, and were
questioned extensively about their smoking and drinking habits
as well as other pertinent aspects of their lifestyles. They were
also given personality tests, since some studies have found an
association between susceptibility to i nfect i ons and personality
type. Blood samples were drawn for analysis, and then most of
the subject s were given nose drops containing an i nfect i ous
dose of the v i rus t hat causes the common cold. (As a check,
twenty-six subject s were given saline drops. As expected, none
of these people got colds.)
To ensure maxi mum control, volunteers were quar ant i ned
at the Cold Unit for seven days af t er being given the nose
drops. During t hi s time, they were monitored daily for cold
symptoms, blood samples were drawn, and a wide range of
other measur es were taken to assess their reactions. Subj ect s
who normally smoked or drank were allowed to continue doing
so throughout t hei r quar ant i ne. Because of the large number
of study subjects and the elaborate measures taken to ensure
accuracy and control, it took nearly three years to conduct the
study.
As the aut hors of the study note in t hei r report, the results
pertaining to alcohol were "unexpected." Contrary to what pre-
vious studies on chronic use of alcohol had led them to expect,
these researchers found t hat alcohol si gni fi cant l y increased re-
sistance to i nfect i on. The volunteers who drank alcohol con-
74 / Buzz
tracted fewer colds dur i ng the st udy period than did the
v ol unt eers who drank no alcohol. And, in general, the more
alcohol consumed, the fewer colds v ol unt eers contracted.
A l i mi t at i on of t hi s st udy is the small number of volunteers
who drank heavily. In fact, only 10 percent of the ent i re sample
and 5.6 percent of the nonsmokers downed more than an av-
erage of three dr i nks dai l y. The aut hor s concluded that t hey
coul dn' t say anyt hi ng about what happens to cold risk as dri nk-
ing becomes heav y. As j us t noted, however, evidence from re-
search on alcoholics indicates t hat heavy dri nkers are more
v ul nerabl e to i nf ect i ons t han are light and moderate drinkers
or abst ai ner s.
As for the mechanism by which alcohol could confer protec-
tion from colds and hel p reduce cold symptoms, there arc only
educat ed guesses at the moment. One possibility is t hat alco-
hol somehow l i mi t s the reproduct i on of vi ruses, either directly
or via an enhancement of the body' s i mmune response. An-
other possibility is that alcohol i nhi bi t s the pr oduct i on of his-
t ami ncst he compounds responsi bl e for r unny noses and
some other unpl easant symptoms of colds. These and ot her
possibilities arc now the focus of ongoing research.
The protective effect s of alcohol seen in this study were not
strong enough, however, to overcome the ef f ect s of smoking.
As expect ed, smokers had the highest i nci dence of colds, and
the more vol unt eers smoked, the more likely they were to catch
a coldregardless of the amount of alcohol they consumed.
As i nt r i gui ng as t hese r esul t s are, much remains uncl ear
about the impact of low doses of alcohol on the i mmune sys-
tem. The aut hors of the st udy end their article with a warni ng
t hat t hei r resul t s should not be t aken as a suggestion t hat non-
dri nkers begin to drink. As with heart disease, the resul t s simply
i ndi cat e t hat those who already dr i nk moderat el y are appar-
ently not increasing t hei r risk of getting a cold. Such people
Sex, Snores, and Stomach Aches I 75
may indeed be giving themselves some small margin of im-
munological protection.
Sex, Lies, and Alcohol
In 1976 f or t y undergraduat e men at Rutgers Uni v ersi t y in New
Jersey set aside their modesty for science. The men took part
in a study designed to shed light on the paradox observed by
Shakespeare that alcohol "provokes the desire, but it takes
away the performance. " Previous researeh had shown that by
most physical measures alcohol is bad for sex. Scientists had
measured how alcohol af f ect ed penile swelling, vaginal en-
gorgement, time requi red to achieve orgasm ( dur i ng both in-
tercourse and mast urbat i on), and vaginal lubrication. The
results were strikingly uni f or m: alcohol inhibited all these re-
sponses. Erections were slower to rise and quicker to fall, va-
ginas were slower to l ubri cat e, and orgasms were slower to
arrive. The mechanics behind these reductions in sexual re-
sponse are still not clearly understood. It is strongly suspected
t hat alcohol inhibits fi ri ng in the peripheral nervous system.
That includes the nerves t ermi nat i ng at the penis, the clitoris,
and the vagina.
The penis and clitoris would respond to t hi s inhibition in
very similar ways because, anatomically speaking, they are
nearly identical except for size. Both st ruct ures contain spongy
t i ssue that can swell and become erect duri ng sexual arousal.
Both penis and clitoris are usually flaccid because the arteries
suppl yi ng these organs with blood are under most circum-
stances clamped tight. Sexual arousal creates nerve impulses
that relax these arteries, allowing blood into the spongy t i ssue
and causing it to swell. Sexual arousal is t hus f undament al l y
dependent on rel axat i on, not tension. Conversely, orgasm in-
volves cont ract i ons of a vari et y of muscles, which j us t goes to
76 / Buzz
show t hat sex is as complex at an anatomical level as it is at a
behavi oral level.
It might seem na t ur a l to assume that since alcohol is a mus-
cle rel axant , it would f aci l i t at e sexnal arousal by relaxing those
al l -i mport ant penile and elitoral arteries. But note t hat the ar-
teries j us t ment i oned r el ax in response to the firing of nerv es,
not the i nhi bi t i on of firing. Those firing nerves can originate
in the brain, as when a person has an erotic f ant asy, or from
the lower spi nal cord, as when the genitals are directly st i mu-
l at ed. The key is t hat anyt hi ng t hat blocks this firing will block
the art eri al rel axat i on needed to achieve an erect penis or clit-
oris. This is exactly how alcohol is t hought to dampen sexual
response, t hough many details in the process remai n obscure.
There arc t i mes, of course, when a l i t t l e dampeni ng is pre-
cisely what a dr i nker is ai mi ng f or . Mast ers and Johnson (1986)
est i mat e t hat between 15 and 20 percent of American men
have at least some di f f i c ul t y controlling premat ure ej acul at i on.
They casually note what is l argel y ignored by most books deal-
ing with sexual problems: t hat many men find t hey can ret ard
overly rapi d ej acul at i on with a j udi ci ous dose of alcohol prior
to sex. Of course, it' s easy to overdo it. If a man dr i nks too
much alcohol, he may well find his sexual response retarded to
the point of impotence.
The ef f ect s of alcohol j us t described, however, tell only a
small part of the story. That ' s because in humans, sex can be
a good deal more complicated t han a mat t er of conjoined gen-
i t al s. The brai n gets i nt o the act as well. And t herei n lies a
t al c.
Despi t e sci ent i fi c evi dence t hat at a purel y physical level
alcohol ret ards sexual response, many people report t hat mod-
erate amount s of alcohol arc good for sex. In one of the largest
surveys addr essi ng the i ssue, 45 percent of men and 68 percent
of women said t hat alcohol enhances t hei r sexual enj oyment
Sex, Snores, and Stomach Aches I 77
(Athanasiou et al. 1970). The answer to t hi s apparent paradox
lies in the old joke about the brain being the body' s most im-
portant sex organ. Especially where alcohol is concerned, t hi s
is no joke at allas the men from Rut gers ably demonst rat ed
in the 1970s. That study involved an ef f or t to separate the
physical ef f ect s of alcohol from the ef f ect s of a person' s belief
about the consumption of alcoholsomething that requi red a
bit of j udi ci ous deception on the part of the scientists (Wilson
and Lawson 1976). It worked like this.
The male st udent s were randoml y assigned to one of f our
groups. One group was given vodka and tonic, and was told
that it was vodka and tonic. Another group got j ust tonic, and
was also told the t r ut h about what they were dri nki ng. The
third and f our t h groups, however, were lied to. The third group
got vodka and tonic, but was told that they were dri nki ng plain
tonic water. (The vodka in these volunteers' dri nks was mixed
in a 1:5 rat i o which was undetectable.) The last group was told
that they were getting alcohol, but in fact they were given tonic
wat er in glasses smeared with a few drops of vodka to produce
an alcohol smell.
The ruses were remarkably ef f ect i v e. Questioned aft er the
experi ment s, not one of the volunteers who. were duped said
they suspected anyt hi ng unusual .
Following the ingestion of t hei r dri nk (ei t her alcoholic or
nonalcoholic), the volunteers were out fi t t ed with a v ar i et y of
monitors to gauge t emperat ure, heart rate, and penile swelling.
Each volunteer then watched an erotic video. Af t er the film,
the v ol unt eers were interviewed extensively and were i nf or med
of the t r ut h if they had been in one of the two groups t hat
were mi si nformed about their alcohol intake.
The resul t s were striking. The subject s who thought they
dr ank alcohol were most highly arousedwhet her they act ual l y
drank alcohol or not. The men who t hought they drank alcohol
78 / Buzz
and who act ual l y got alcohol were the most hi ghl y aroused. The
men who thought they got alcohol but got only tonic water
were slightly less aroused, but these men were si gni fi cant l y
more aroused t han those who expected tonic water but act ual l y
drank alcohol. Thus it was t he belief in alcohol consumpt i on
that proved si gni fi cant to sexual response, not the presence or
absence of alcohol. The belief overcame any of the physiolog-
ical dampeni ng ef f ect s that the alcohol might have had.
These findings have not been the onl y ones showing that
belief and expectations arc more i mport ant than purely phys-
ical ef f ect s of alcohol. A similar st udy f ound t hat people who
t hought they drank alcohol were si gni fi cant l y more aggressive
in a social si t uat i on than people who t hought t hey drank tonic
wat erregardl ess of whet her they act ual l y did or did not dri nk
alcohol (Lang et al. 1975). More recent st udi es have shown
that belief similarly ef f ect s f emal e sexual response and the re-
sponse of sexual l y i nhi bi t ed males (Lang et al. 1980). When it
comes to alcohol, in other words, people of t en feel what they
expect to feel . This process can obviously be sel f-rei nforci ng.
The experience of an enhanced sexual encount er under the
i nf l uence of alcohol can lead to increased expect at i ons of sim-
i l ar results t he next time ar ound.
One recent l y report ed i nt eract i on between alcohol and sex
deserves ment i on, if only because it has been so badly ex-
plained in the popul ar press.
In a 1995 st udy, Finnish and Japanese researchers who were
st udyi ng some metabolic aspect s of alcohol consumpt i on un-
expectedly discovered a rel at i onshi p between alcohol consump-
tion and testosterone levels in women. They f ound that the
or di nar i l y low t est ost erone levels in women rise dramat i cal l y
one to two hours af t er women imbibe alcohol, and t hat the
rise was most dramat i c in women who were ei t her ovulating or
Sex, Snores, and Stomach Aches I 79
taking oral contraceptives. The finding was published in the
staid scientific j our nal Nature, but it was picked up by many
considerably less-reserved media out l et s because testosterone
has been shown to increase sexual desire in both men and
women. Many media report s suggested t hat women might thus
respond more fav orabl y to sexual advances a f t e r a few dri nks.
This is qui t e a leap f r om the v ery car ef ul l y drawn findings
of the study. For one thing, the findings have yet to be repli-
cated. More i mport ant , it r emai ns to be seen whether the levels
of testosterone detected in the st udy have anyt hi ng to do with
actual behavior. There are good reasons, in f act , to suspect that
they do not. First of all, the study simply measured total tes-
tosterone in the blood of female volunteers. It t ur ns out that
most of a woman' s testosterone is not biologically activeit is
bound to a prot ei n in the blood and does nothing. Whether
alcohol ingestion act ual l y raises the level of biologically active
testosterone is unknown. And, as we've j ust seen, the beliefs
and expectations about alcohol's ef f ect s are likely to be more
powerful i nf l uences on behavior than any ef f ect exerted by tes-
tosterone.
It is i mport ant to remember t hat these resultsas well as
all the r esul t s showing the impact of expect at i on on sexual
responseare based on moderate dr i nki ng. The blood alcohol
levels among the vol unt eers in the expectancy st udi es, for in-
stance, were equivalent to what would be f ound af t er the con-
sumption of only two or three standard dri nks over the course
of an hour. (Again, a st andard dr i nk is defined as a half-ounce
of pure alcoholthe amount generally f ound in a 12-ounce
can of beer, a 5-ounce glass of wine, or a 1.5-ounce shot of
whiskey.) If higher doses had been used in the experi ment s, it
is likely that all the physical variables measured, including pe-
nile swelling, would be adversely af f ect ed. There are l i mi t s, in
80 / Buzz
ot her words, to mi nd over mat t er . Given a hi gh enough dose,
all the expect at i on in the world won' t rouse a penis or clitoris
anest het i z ed by alcohol.
Indeed, the h a r mf u l ef f ect s of chronic dr i nki ng have been
so exhaust i v el y chroni cl ed t hat i t ' s easy to see why those in
posi t i ons of i nf l uence are caut i ous in t hei r comments on such
a mat t er as the pot ent i al usef ul ness of low closes of alcohol for
pr emat ur e ej acul at or s. A br i ef list of such consequences will
s uf f i ce to i l l us t r at e the poi nt . Chroni c alcohol use can quash
the l i bi do of both men and women. It can seri ousl y depress
testosterone levels in males, it of t en causes a pronounced
shr i nki ng of the testicles, and it st rongl y i mpai r s the ability to
achieve and sustain an erection. In women, chronic use of al-
cohol reduces vaginal response, and it can cause i rregul ar men-
s t r uat i on and induce pr emat ur e menopause.
The Exception to the Rule
Related to the issues of alcohol and sex is the mat t er of alco-
hol's di re ef f ect s on a developing f e t us . Although some doctors
cont i nue to believe that a very occasional dr i nk by a pregnant
woman is harmless, the t i de of medi cal opinion on t hi s mat t er
has shi f t ed in recent years in light of new research. For in-
stance, al t hough it has long been known t hat alcohol passes
qui ckl y and easily across the pl aeent al bar r i er between mother
and f et us, it now appears t hat alcohol may af f ect an embryo
even bef or e it has i mpl ant ed i t sel f in the ut er i ne wall and be-
come engaged wi t h t he mother' s blood (Coles 1994). Much
research is now aimed at discovering the exact mechani sms by
which alcohol may har m a developing f et us . More t han likely,
alcohol i mpai r s many cri t i cal molecular syst ems at once.
One system now being scrut i ni z ed is absol ut el y essent i al for
t he proper wi r i ng up of neur ons in the br ai n of the devel opi ng
Sex, Snores, and Stomach Aches I 81
f et us . Special molecules called adhesion mol ecul es guide the
migration of developing neurons and help them to make stable
connections to ot her neurons. Michael Charness and his col-
leagues at the Harv ard Medical School demonst rat ed that
alcohol "st ri ki ngl y reduces" the ability of cert ai n adhesion mol-
ecules to promote the format i on of stable mul t i eel l ul ar orga-
ni z at i ons (Charness et al. 1994). The di sr upt i v e ef f ect s of
alcohol on cell adhesion molecules is a specific exampl e of the
havoc alcohol is t hought to wreak on many ot her molecular
mechanisms critical for the proper development of a f et us .
Although it takes very heavy dr i nki ng indeed (an average
consumpt i on of fort y-t wo st andard dr i nks a week) to produce
the severe physical deformities and mental retardation char-
acteristic of f et al alcohol syndrome, no saf e threshold has been
f ound for t he far more subtle and di ffi cul t -t o-measure ef f ect s
of alcohol on mental development. To quote f r om a recent
st udy of the mat t er : "For some behaviors, such as mental de-
velopment, even the smallest prenat al dose of alcohol appears
to have some adverse ef f ec t on the f et us, and the severity of
the ef f ect increases gradual l y with i ncreasi ng levels of expo-
sure" (Jacobson and Jacobson 1994).
This i nabi l i t y to det er mi ne a safe level of dri nki ng has led
many doctors and public health of f i ci al s such as the U.S. sur-
geon general to advise women who are either pregnant or t ryi ng
to get pregnant to abst ai n completely from alcohol. And it' s
probably not a bad idea for the male to qui t dri nki ng tooat
least while trying to f at her a child. New evidence shows t hat
alcohol can have an adverse ef f ect on sperm, which may induce
subtle yet marked defi ci t s in the of f s pr i ng of alcohol-exposed
fat hers (Cicero 1994).
In sum, if you are hav i ng sex with the i nt ent i on of conceiv-
ing a child, most cur r ent research suggests t hat you stick to
sparkling cider and other nonalcoholic dri nks. If you' re simply
82 / Buzz
having sex for the pl easure of it, the research i ndi cat es that
moder at i on is pr udent . General l y speaking, the impact of al-
cohol on sexual response is dose-dependent. At light to mod-
erate levels, the brai n is more i mpor t ant t han t he alcohol: the
response you f eel will have more to do with what you think
you will feel t han with the pharmacological impact of alcohol
on your sex organs. That means alcohol can ei t her hel p or hur t
sex, dependi ng on your expect at i ons. As alcohol consumpt i on
increases so does its power to override the mind and di rect l y
dampen sexual response. And, at high doses, as Shakespeare' s
Porter observed, you may well find your desire "provoked," but
you will most l i kel y be robbed of your per f or mance.
The Morning After
Hangovers are the bane of dr i nker s. The throbbing head, nau-
sea, i r r i t abi l i t y, dry mout h, l et hargy, and hypcr scnsi t i v i t y t o
l i ght and sound make t hi s condi t i on so unpl easant t hat most
people try to avoid it at all costs. Aside f r om the obvious tack
of dr i nki ng slowly and in moderat i on, t her e arc i nnumerabl e
fol k remedi es aimed at avoiding or curing hangovers. Most of
these ideas are i l l - f ounded, some are downri ght har mf ul , and
a few act ual l y provi de some r el i ef .
The poundi ng headache common to hangov ers has two pos-
sible sources. First of all, as Shakespeare' s Porter again pointed
out, alcohol is a di ur et i ct hat is, it promotes ur i nat i on. It does
t hi s by blocking an i mpor t ant substance in the kidneys called
ant i di ur et i c hormone, or ADH, which a dj us t s t he porousness
of the microscopic tubes t hat car r y ur i ne out of the ki dney.
Normal l y, most of t he wat er in ur i ne is recycled t hrough the
porous walls of the collecting t ubes. But when ADH release is
blocked by alcohol, the t ubes become less porous, t hus cut t i ng
Sex, Snores, and Stomach Aches I 83
down on rcabsorption and increasing ur i ne out put . (As we'll
see later, caf f ei ne also promotes uri nat i on, but in an ent i rel y
di f f er ent way. ) Somewhat ironically, t hen, dr i nki ng large
amount s of alcohol can lead to mild dehydration. This, in t ur n,
can lead to both a dry mouth and a headache owing to reduced
blood pressure in the cranial vessels. The second way alcohol
can induce a headache is by relaxing and enlarging the same
vessels in the headan action that compounds the low-blood-
pressure problem created by dehydrat i on.
Both of these problems are, to a certain ext ent , correctable.
Downing plenty of wat er both duri ng and af t er dri nki ng can
help prevent dehydrat i on (though it may add to a restless night
by increasing your need to go to the bathroom). Drinking a
caffei ne-cont ai ni ng beverage in the morni ng may help also be-
cause caf f ei ne const ri ct s cerebral blood vessels. Taking an
aspirin or two may help also, though only by maski ng the pain,
not by solving the basic problem. Taki ng aspi r i n before you
start dri nki ng is not a good idea. As noted in Chapt er 2, aspi ri n
i nt er f er es with alcohol dehydrogcnase, which can lead to higher
blood alcohol levels and worse hangover symptoms.
Another common hangover symptom is a general l et hargy
and muscul ar weakness. One fact or cont r i but i ng to t hi s malaise
is the bui l dup of lactic acid in the muscles t hat can fol l ow
heavy dri nki ng. As most athletes know, the f at i gue and cramp-
ing caused by st r enuous exercise result f r om the accumul at i on
of lactic acid and a subsequent di srupt i on of the acid-base
balance in the muscles. Drinking alcohol can do roughly the
same thing, though by a di f f er ent rout e. The enz ymat i c de-
st ruct i on of alcohol in the liver requi res many i mport ant
"helper" molecules. Buf these helper molecules are normal l y
used to process many other t oxi ns, i ncl udi ng lactic acid. When
present ed wi t h a load of alcohol, the liver and all its en/ ymat i c
84 / Buzz
machi ner y drop what t hey' re doing and go to work on the al-
cohol. This allows less dangerous t oxi ns like lactic acid to ac-
cumul at e, creat i ng overly acidic conditions in your muscles.
The body's acid-base bal ance can be thrown off anot her way
as well. The pri nci pal product of alcohol met abol i sm is acetic-
acid, which is us ef ul in many ways. But produced in excess
f r om the breakdown of alcohol, it can simply aci di f y the blood,
exacerbating the lethargic feel i ngs produced by the l act i c acid
bui l dup. There is l i t t l e one can doother t han wai t t o cor-
rect such imbalances. Mild exercise may help a l i t t l e by in-
creasing blood ci r cul at i on and t hus flushing lactic acid from
the muscles. But t hi s strategy could easily backfi re, since stren-
uous exercise would simply produce more lactic acid, maki ng
the si t uat i on worse.
The queasy stomach common to hangovers is of t en at t ri b-
utable to the i ncreased acids secreted by the stomach in re-
sponse to alcohol. Of course, ot her fact ors may be at work as
well, not least of which could be episodes of vomiting, which
would leave the stomach both empty and overly acidic. Eat i ng
a light carbohydrat e such as toast, crackers, or cereal is of t en
recommended in this si t uat i on because it helps neut ral i z e acid
and is easily digested.
Hypersensi t i v i t y to light or sound may be due to the "re-
bound" ef f ect discussed earl i er in t hi s chapter. A heavy bout
of dr i nki ng will produce t emporary withdrawal symptoms as the
brain and body st ri ve to rebalance themselves. Since with-
drawal generally produces sympt oms t hat are the ant i t hesi s of
the ori gi nal ef f ect s of a part i cul ar drug, the rebound from al-
cohol of t en brings with it increased excitability, depressed
mood, and sensi t i v i t y to st i mul i .
The fact that a hangover is, in par t , a drug-withdrawal symp-
tom accounts for the long-standing "hair-of-the-dog" cure.
(The phrase comes f r om an old Bri t i sh saying: "A hair of the
Sex, Snores, and Stomach Aches I 85
dog that bit you"a metaphor for the idea that a small
amount of the same ki nd of l i quor drunk to excess the previous
night will cure a hangover. ) This "cure" can act ual l y work
t emporari l y. By re-creating the conditions to which the brain
and body had become accustomed dur i ng the night of drink-
ing, a "hair-of-the-clog" nip of alcohol can bring some rel i ef.
But, of course, this only postpones the final reckoning and leads
to more intense wi t hdr awal symptoms l at er on. Fortunately,
many people are dissuaded from a "hair-of-the-dog" because
t hey have a nat ur al aversion to alcohol following a significant
pub crawl.
iMany people feel t hat di f f er ent kinds of dr i nks produce di f-
ferent kinds of hangovers. From a purel y theoretical point of
view, there is some logic to t hi s. Wines, l i quors, and beers
contain hundr eds of complex molecules that give these dr i nks
t hei r charact eri st i c flavors, smells, and appearances. Such com-
pounds are collectively called "congeners." Generally speaking,
the more congeners a dr i nk has, the darker it will appear and
the richer it will t ast e. Red wine, for instance, has more con-
geners than white wine. Scotch, cognac, and brandy have more
congeners than gin, and gin has more congeners t han vodka,
which is arguably the most congener-free l i quor of all.
The problem with congeners is that there are so many of
them that nobody has gone to the trouble of testing to see
what , if any, ef f ect t hey have on either i nt oxi cat i on or hango-
vers. (A few congeners, as mentioned, have been tested, but for
their ef f ect s on the cardi ovascul ar system, not on hangovers. )
One congener wi t h proven abilities to contribute to intoxica-
t i on is methano], ethanol' s simpler cousi n. Methanol isn' t a
direct product of f er ment at i oni t is probably derived from the
breakdown of pectins in f r ui t - based wines or liquors. Red wine
has more met hanol than white wines, but even the amount in
red wine is so small (less t han 1 percent by vol ume) that it is
86 / Buzz
unl i kel y to play a role in hangovers. Other congeners i ncl ude
dozens of phenols, t anni ns, sul f ur - cont ai ni ng compounds, or-
ganic acids, ami no acids, esters, sugars, and gasses such as car-
bon di oxi de. In all, more than five hundred di st i nct kinds of
congener molecules have been i dent i fi ed in wine alone (Amer-
ine and Roessler 1983).
It is possible t hat some people are more sensitive to the
congeners in cert ai n dri nks t han in others, and thus find t hat
those dr i nks (such as red wine) give them worse hangovers. It
is equal l y pl ausi bl e t hat congeners have not hi ng to do with
hangovers, and any appearance to the cont rary is due ei t her to
expect at i ons on the part of the dr i nker or si mpl y to the very
real ef f ec t s pr oduced by the alcohol.
The Mi ddle Road
In this chapt er we've explored some of alcohol' s ef f ect s on the
human body. We've seen that moderate amount s of alcohol,
par t i cul ar l y red wine, conf er protection agai nst heart disease
though doctors don' t recommend that those current l y abstain-
ing begin dri nki ng to secure these modest benefi t s. We've seen
t hat alcohol can at first promot e sleep, and then di st urb it by
i nduci ng a st i mul at i ng "rebound" ef f ect . We've looked into
the age-old idea t hat a measured dose of alcohol helps speed
heal i ng or prev ent si ckness and f ound a grain of t r ut h in an
otherwise unprov ed not i on. We've been reminded that the
most i mpor t ant organ in sexual response is the brai n, and we've
seen t hat alcohol' s i nhi bi t i ng ef f ec t s on the sex organs become
increasingly i mpor t ant as the dose increases. And finally we've
explored the miseries of hangov ers and seen how little can be
done to prevent one, other t han dri nki ng moderately in the
first place.
Indeed, moderat i on has been a repeated r ef r ai n in this chap-
Sex, Snores, and Stomach Aches I 87
t er. The research to date on the wide-ranging ef f ect s of alcohol
on the human bodyboth positive and negativeindicates
that for moderate dri nki ng, the net health impact is mi ni mal
or even slightly benefi ci al . Moderate drinking, remember, is
usually defined as no more than two standard dri nks a day for
men and no more than one standard drink for women (Gordis
et al. 1995). Remember, too, that this rule of t humb doesn' t
apply to pregnant women or to men and women who are trying
to conceive a child.
And the biggest caveat, by f ar , to the rule of moderation is
that all of the findings presented here apply only to people not
at risk for alcoholism.
The best current estimate is that roughly one in ten drinkers
is alcoholic. That means t hat for one in ten drinkers, the con-
cept of moderate drinking is a dangerous illusion. Although
some therapy programs claim that some alcoholics can resume
moderate, controlled drinking, there remains great debate over
this suggestion and subst ant i al doubt about its practical im-
pl i cat i ons.
It is to the subset of drinkers for whom alcohol is powerful l y
addicting that we now t ur n. No other area of alcohol research
is as charged with emotion, contention, and debate. And yet
the recent discoveries here are among the most f asci nat i ng in
the entire field.
Of Mi ce and Men
Since you can' t open up a person' s skull and probe ar ound a f t er
t hey' v e had a dr i nk to see what ' s happeni ng, sci ent i st s who are
cur i ous about such t hi ngs use ani mal s. Rats and mice arc by-
far t he f av or i t e cr eat ur es for alcohol research. They' re rel at i v el y
i nexpensi v e, and t hei r br ai ns arc a lot like huma n br ai ns, only
smal l er.
But as valuable as ani mal s are, they have a seri ous drawback:
by and large, t hey hate alcohol. When alcohol s t udi es usi ng
ani mal s began in earnest in the 1950s, it was f ound t hat when
Demon Rum I 89
dogs, cats, pri mat es, rat s, and mice were given a choice be-
tween an alcohol solution and water they almost invaribly
chose wat er. This posed qui t e a problem for researchers who
want ed to see how alcohol i nt akepar t i cul ar l y long-term in-
t akeaf f ect ed the brain and other body systems.
Researchers overcame t hei r subj ect s' nat ur al aversion to al-
cohol in a v ari et y of ways. They admi ni st ered alcohol i nt r av e-
nously, for instance, or they disguised the alcohol with sugar
or sugar subst i t ut es. Some sci ent i st s even filled the ani mal s'
cages with vaporized alcohol, t hus using the respi rat ory system
as a means of alcohol ingestion. As awkward as such methods
sound, they nonetheless allowed researchers to learn a great
deal about how alcohol works. Still, humans obviously don' t
inhale their dr i nks (at least not l i t er al l y), and they usual l y dri nk
vol unt ari l y. Indeed, the si t uat i on that many researchers were
most interested i ndel i berat e, chronic dr i nki ngwas the one
most di f f i cul t to accurately model with nat ur al l y abstinent lab
ani mal s.
Then, in the 1950s, sci ent i st s in Chile, Fi nl and, and the
United States i ndependent l y succeeded in breeding st rai ns of
mice that act ual l y liked alcohol (Crabbe et al. 1994). In 1959
the team of Gerald McClearn and David Rodgers at the Uni -
versity of Cal i forni a, Berkeley, discovered that a strain inglori-
ously called C57BL/Crgl clearly preferred alcohol to water when
given a choice (McClearn and Rodgers 1959). They drank so-
l ut i ons with an alcohol content of 10 percent roughl y the
same as white wine, and strong enough to make t hem t i psy.
The discovery was welcomed by experi ment al i st s. They now
had an animal t hat , like some humans, pr ef er r ed alcohol to
wat er. But the existence of alcohol-drinking mice was more
pr of ound t han a mere methodological breakt hroughi t st ruck
to the heart of a central issue in alcohol studies. These mice
were impossible-to-deny evidence t hat a preference for alcohol
90 / Buzz
could be due to genes. The C57BL/Crgl mice were not some
exotic species of mouse with radi cal l y di f f er ent physiologies
f r om ot her mice. They were si mpl y one of six car ef ul l y bred
f ami l i es of Mus musculuscommon house micetested by
McClearn and Rodgers. The si t uat i on is analogous to selecting
a few members of six unr el at ed human fami l i es, giving them
two unlabeled gl asses of l i qui d, letting them sample both, and
f i ndi ng t hat the members of one f ami l y pr ef er r ed t hei r wat er
laced with alcohol.
The mice' s preference was compelling because it clearly had
not hi ng to do with upbri ngi ng, cul t ur e, peer pressure, stress,
expect at i ons, adv ert i si ng, emotional t r auma, or any ot her var-
iable t hat can i nf l uence human dr i nki ng. The preference shown
by the C57BL/Crgl mice was internally generat ed. They inher-
ent l y liked the taste, the intoxication, or some other qual i t y of
the alcohol. Subsequent breedi ng bore this out. When alcohol-
pr ef er r i ng mice were bred together, the inborn predilection
st rengt hened: the grandchi l dren of the original mice drank
more and dr ank hi gher concent r at i ons of alcohol than t hei r
gr andpar ent s. Likewise, when mice that avoided alcohol were
bred, t hei r progeny became less willing to take even a sip of
t he hard s t u f f .
When these exper i ment s were conducted, the science of ge-
netics was in its i nf ancy. Nobody knew what was going on in-
side those mice, though it was clear t hat the answer could be
f ound somewhere in the tight coils of t hei r DNA. Today, of
course, genetics is one of the hot t es t fields in science. Genes
now can be "read" wi t h r el at i v e ease, and t hi s new abi l i t y is
rev ol ut i oni z i ng our under st andi ng of both physical and ment al
illnesses. Specific genetic defect s have been f ound responsible
for Hunt i ngt on' s disease, sickle-cell anemia, cystic fibrosis,
muscul ar dystrophy, and a host of other diseases. And since
DNA directs the const r uct i on of brai ns as well as bodies, gc-
Demon Rum I 91
netic v ari at i on is coming to be seen as a key player in people's
ment al makeup.
This new appreci at i on for the way genes can subtly i nfl uence
things l i ke personal i t y and mood has cont ri but ed to a sea
change in the popular view of alcoholism. Alcoholism was long
thought to be caused by a f ai l ur e of will, a lack of moral fiber,
or simple i rresponsi bi l i t y. The pendul um of public opinion has
in recent years swung in the opposite direction: most people
now view alcoholism as a disease caused by a genetic mal f unc-
tion t hat renders s uf f er er s predisposed to abuse alcohol. A 1987
Gallup poll found that nearly 90 percent of Americans believe
that alcoholism is a disease, and more t han 60 percent think
that it may be inherited.
Evidence to support t hi s model of alcoholism has been ac-
cumul at i ng for decades. The discovery of those alcohol-
pr ef er r i ng mice in the 1950s was one of the early cont ri but i ons
to what is popul arl y known as the "disease model" of alcohol-
ism. The heady ent husi asm generated by the early findings of
genetic predispositions to alcoholism, however, has faded and
been replaced by a growing appreci at i on for the l i mi t at i ons of
a purely biological approach to problem drinking. For one-
thing, despite f or t y years of searching, nobody yet knows why
those mice discovered in the 1950s like alcohol. Whatever the
genetic di f f er ence is between the t i ppl i ng mice and their non-
dri nki ng bret hren, it is so subtle or complicated t hat it has yet
to be i dent i fi ed. But even if the genetics responsible for the
mice' s preference were i dent i fi ed, it is not clear t hat those find-
ings would really shed much light on human dri nki ng. As sim-
ilar as mouse br ai ns are to human brai ns, mice are not men.
It t ur ns out t hat t he human brai n' s capacity t o generate things
like beliefs and v al ues can rival the power of genes to i nfl uence
drinking behavior.
Wi t hi n the field of alcoholism research there is a growing
92 / Bu z z
appreci at i on for t he subt l et y of t he di sor der f uel ed pr i mar i l y
by the fact t ha t v i r t ual l y all the evidence amassed to support
a role of genes in alcoholism i ndi cat es t hat f aul t y genes cannot
be the only cause of alcoholism. Cul t ur e, f ami l y env i r onment ,
l ear ni ng, st ress, even the hoary old notion of wi l l power, can
pl ay i mpor t ant roles as wel l . In other words, it is no longer
nat ur e versus nur t ur e . I t ' s nat ur e and n u r t u r e . Or, perhaps
more accur at el y, nat ur e/ nur t ur enot t wo separat e ent i t i es,
but two si des of the same com.
In t hi s chapt er we will survey t hi s new and still-evolving con-
ception of alcoholism. We'll see what is known about why mice
and men v ar y in t hei r propensi t y to dr i nk. In the process, we' ll
see that the once seemingly elcar l i ne between "alcoholie" and
"nonalcoholic" has become qui t e bl ur r y. No longer is it si mpl y
a mat t er of having or not hav i ng a disease. The forces at work
on an alcoholic are at work, to one degree or anot her, on every-
one, i ncl udi ng the abst i nent . The sci ence of alcoholism is in-
exor abl y l eadi ng toward a v i ew of the problem t ha t is more
compl i cat ed, more human, and more honest t han ci t her of the
pol ar ext r emes that have eharaet eri / . ed the debate for cent u-
ries.
Papa's Legacy
Alcoholism tends to run in f ami l i es . The prevalence of alco-
holism in the general popul at i on of males is est i mat ed to be
between 3 and 5 percent, whi l e the prevalence among male
rel at i ves of alcoholics is about 25 percent (Goodwin 1985). For
f emal es, the figures are much lower, though the trend is sim-
i l ar . The prevalence of f emal e alcoholism in the general pop-
ul at i on is between .1 and 1 percent, while the prevalence
among femal e rel at i ves of alcoholics is est i mat ed at between 5
and 10 percent .
Demon Rum I 93
These numbers say not hi ng about the causes of alcoholism.
Many things in addi t i on to genes get passed from generat i on
to generat i on, among them learned behaviors such as might
account for a tendency to dri nk. The numbers j ust cited also
strikingly cont radi ct the notion that chi l dren of alcoholics are
somehow destined to become alcoholics themselves. Although
having a close alcoholic rel at i ve (parent or sibling) clearly in-
creases one's risk of alcoholism, it is equally clear that hav i ng
such a rel at i ve docs not, by any means, condemn one to al-
coholism. Seventy-five percent of males with an alcoholic rel-
ative do not become alcoholics themselves; more recent figures
put the figure closer to 80 percent. Between 90 and 95 percent
of women with an alcoholic relative escape the disorder. If al-
coholism is a disease, in other words, it either is inherited in a
most pecul i ar manner or is so weak t hat most people manage
to overcome it.
One way researchers have t ri ed to tease apart the entwined
st rands of genetic and env i r onment al i nfl uences on alcoholics
is to study t wi ns. There are, of course, two ki nds of t wi ns.
Identical twins f or m from a single egg and t hus share identical
genes, while f r at er nal twins are derived f r om two separat e eggs
and are no more genetically si mi l ar t han non-twin siblings. If
alcoholism t r ul y has a genetic component, then identical twins
should tend to develop more si mi l ar dri nki ng pat t er ns and
problems t han f r a t er na l twins.
Si mi l ari t y in a t rai t is measured with a value called concor-
dance. If twins are identical in a t r ai t eye color for instance
the concordance is 100 percent . If they are completely
divergent in a t r ai t , the concordance is zero. Identical twins,
clearly, are far more concordant in general than are f r at er nal
t wi ns.
The resul t s of numer ous st udi es from around the world
clearly show that both genes and the env i ronment i nfl uence
94 / Buzz
dr i nki ng behav i or. As many suspected, all of the st udi es have
found t hat i dent i cal twins share the t r ai t of alcoholism or prob-
lem dr i nki ng more of t en t han ei t her f r at er nal twins or com-
pletely unrel at ed people ( Depart ment of Health and Human
Services 1993). For i nst ance, a recent st udy of female twins
f ound t hat the heri t abi l i t y of alcoholism was between 50 and
60 percent (Kendl er et al. 1992). The aut hor s concluded that
"genetic f act or s play a maj or role in the etiology of alcoholism
in women." But the magnitude of the di f f er ences observed in
the heri t abi l i t y of alcoholism in t wi n st udi es is oft en surpri s-
ingly modest. A st udy of identical male twins showed a con-
cordance rat e for alcohol abuse or dependence of 76 percent,
while that of f r at er nal twins was 61 percent. Although this is
a st at i st i cal l y si gni fi cant number , it' s har dl y a ringing af f i r ma-
tion of the disease model. The figures for women were even
less impressive: a 36 percent concordance for i dent i cal twins
compared with a 25 percent concordance for f r at er nal twins
(Pickens et al. 1991).
This is an example of how a single st udy can support dia-
met ri cal l y opposed views: those who favor genetic i nfl uences
can point to the di f f er ences in concordance as proof, while
those who t hi nk env i ronment al condi t i ons play a bigger role
note that if alcoholism were purel y a mat t er of genes, then the
concordance for i dent i cal t wi ns should be close to 100 percent.
That it' s not indicates that env i ronment al variables are at work.
In real i t y, few people argue t hi s way. Although it is a point
of t en lost in lay di scussi ons of alcoholism, pract i cal l y nobody
in the field believes t hat the disorder resul t s from ei t her genes
or the env i r onment alone. Even the most die-hard champions
of genetics acknowledge t hat no amount of genetic predispo-
sition can i nduce alcoholism if, for exampl e, no alcohol is avail-
able for consumption. Likewise, even those who t hi nk
upbri ngi ng or some other env i r onment al fact or is key to alco-
Demon Rum I 95
holism admit that for at least some people a genetically based
vulnerability may play a role. The one thing that seems clear
at least to scientistsis that alcoholism is not the result of a
single f aul t y gene such as that underl yi ng sickle-cell anemia or
Huntington' s disease. Most researchers tracking down the ge-
netics of alcoholism and other types of addiction now assume
that these t rai t s spring from the combined i nfl uence of several
genes, not one.
Some of the most i nt erest i ng evidence to support this idea
comes from studies on those alcohol-preferring mice discovered
in the 1950s.
Mouse Tales
If you wanted to, you could pick up the phone today and order
a mouse or a rat that displays any of more t han a dozen distinct
reactions to alcohol. You could, for instance, buy a mouse that
becomes sleepy and motionless af t er a modest dose of alcohol.
Such mice are called long-sleep (LS) mice because it takes
them an unusual l y long time to wake up from an alcohol-
induced nap. Conversely, another strain of mice gets drowsy
on the same dose of alcohol, but quickly ret urns to normal.
Hence, they' re called short-sleep (SS) mice.
Or, if you' d prefer, you could choose between a COLD
mouse and a HOT mouse. The metabolism of COLD mice
slows down following a dri nk of alcohol, while the metabolism
of HOT mice speeds up. There are also mice that become
energized and active af t er dri nki ng alcohol, and their opposites
that become lethargic. There are even two st rai ns that exhibit
opposite sensitivities to withdrawal symptoms: a strain that dis-
plays the tremors and seiz ures typical of humans in the throes
of delirium tremens (DTs) and a st rai n resistant to such symp-
toms.
96 / Buzz
The moral of this st ory is t hat , to a cert ai n ext ent at least,
the di f f er ent physiological aspects of i nt oxi cat i on can be selec-
tively bred i nt o or out of ani mal s . The anest het i z i ng aspect can
be separat ed from the s t i mul at i ng aspect, for i nst ance, or sen-
s i t i v i t y t o wi t hdr awal can be separat ed f r om t he t hermal ef f ec t s
of alcohol i ngcst i on. It appear s, t her ef or e, t hat i ndependent
genetic f act or s i nf l uence alcohol s ens i t i v i t y, tolerance, depend-
ence, and s el f - admi ni s t r at i on and t hat these t r ai t s are sup-
ported by di st i nct neurobiological mechani sms (Crabbe et al.
1994). This may be one reason for t he di v er s i t y in t he experi-
ences of dr i nker s. With an unknown number of genes at work
helping shape one' s sensi t i v i t y to alcohol, it' s not sur pr i si ng to
sec v ari at i ons in people' s responsesand that doesn' t take into
account the even great er v ar i at i ons produced by people' s d i f -
f er ent personal i t i es and ot her t r ai t s .
Some at t empt s to categorize the di f f e r e nt responses to al-
cohol have been made. The most widely accepted distinction
is between type I and type II alcoholism. Type I is more com-
mon, appears in both men and women, is less severe t han t ype
II, and of t en appear s in mi dl i fc rat her than early on (Cloninger
et al. 1981). Several st udi es suggest t hat t hi s type of alcoholism
has a less pronounced genetic component and that its expres-
sion is st rongl y dependent on env i r onment al i nf l uences. Type
II alcoholism, in contrast, is charact eri z ed by a severe suscep-
t i bi l i t y t hat seems to be expressed regardless of the environ-
ment . Type II al cohol i sm occurs onl y in men, develops early
( of t en in adol escence), and is much more di f f i cul t to treat t han
type I al cohol i sm.
Clomnger' s division of alcoholism i nt o two types is not the
l ast word on the mat t er . Ot hers have suggested a third type:
alcoholism ari si ng f r om a pri mary ant i soci al personality or
mood di sorder t hat exi st s pri or to drinking (Schuckit 1985).
The ext ent to whi ch these categories reflect under l yi ng genetic
di f f er ences i s unknown. I f , as the ani mal st udi es suggest, as-
Demon Rum I 97
pects of intoxication such as craving and withdrawal sensitivity
are regulated by di f f er ent genes, then people with defects in
such genes could plausibly express di f f er ent forms of the dis-
order.
This idea is simply a hypothesis at the moment, though it
is being t aken very seriously by a number of researchers. The
animal st udi es arc so t ant al i z i ng t hat many t eams of scientists
are s i f t i ng through animal genomes looking for the ki nd of
def i ni t i v e defect or al t erat i on they suspect lies at the bottom
of the behavioral v ari at i ons that can be bred in or out. Re-
cently, they' ve had some good luck. The resistance to alcohol
shown by short-sleep mice has been linked to a t i ny v ari at i on
in the gene used to build a type of GABA receptor in mouse
brains. This v ari at i on, which slightly changes the st r uct ur e of
the receptor, is appar ent l y vital to making the receptor sensitive
to alcohol. Without t hi s piece, alcohol can' t al t er the GABA
receptors as it usual l y does, thus leaving the mice rel at i vel y
resi st ant to i nt oxi cat i on.
As exciting as this ki nd of discovery is, it' s a long way from
a definitive statement about how genes af f ect human dri nki ng
behav i or. Thus far researchers have been unable to find a hu-
man equi v al ent of the mouse GABA v ar i ant , much less a gene
responsible for alcoholism in general. Thus despi t e widespread
popular belief in the genetic origins of alcoholism, the sci ent i fi c
j ur y is still far from reaching a verdict in the ease. Nothing
illustrates t hi s fact so eompellingly as the story of what was, at
one time, the leading candi dat e for the put at i v e "alcoholism
gene."
D
2
or Not D
2
?
It began with a highly publicized press conference arranged to
coincide with the appearance of the April 14, 1990, issue of
the Journal of the American Medical Association cont ai ni ng an
98 / Buzz
article by Kennet h Blum, a pharmacologist at the Uni v er si t y of
Texas, and Ernest Noble, a psychi at ri st at the Uni v ersi t y of
Cal i f or ni a. The pai r announced t h a t t hey had i dent i f i ed a ge-
net i c defect in the dopami nc syst em of some alcoholics. Do-
paminc, as we saw earlier, is a key component of the brai n' s
reward pat hwaysi n many ways, it is the neur ot r ansmi t t cr of
pl easure.
Bl um and Noble exami ned DNA samples from the br ai ns of
corpses of t hi r t y- f i v e alcoholics and t hi r t y- f i v e nonaleoholics.
They f ound t hat a v ar i ant of the gene for a specific type of
doparnine recept or (D
2
) was present in 69 percent of the al-
coholics but only 20 percent of the nonal eohol i cs. They
claimed that this variant, called the Al allele, results in fewer
D
2
recept ors in the brain (Noble et al. 1991). Fewer receptors
could lead to an i mpai red reward syst em. In short, the team
t heori z ed, people i nher i t i ng this gene might get less pl easure
less of an i nt er nal "hi gh"from enj oyabl e l i f e events. When
such people dr i nk dopami nc-boost i ng alcohol, they might feel
"normal" for the first t i me in t hei r lives. Such a feel i ng would
be very powerful and could lead to the i nt ense craving observed
among alcoholics.
It was a beaut i f ul t heory, and it was f ai r l y easy to under st and.
Not sur pr i si ngl y, the story made the front page of the New York
Times and many other paper s. It was hai l ed in popul ar media
as the ul t i mat e confi rmat i on of the disease theory of alcohol-
ism and the humbl i ng of the backward "nurt uri st s, " who claim
that alcoholics arc, to some ext ent anyway, responsible for t hei r
own condition.
U nf or t unat el y, pl ausi bl e though the D
2
t heory is, it hasn' t
been embraced by t he alcohol research communi t y (Holdcn
1994). Six months a f t e r Blum and Noble' s paper appeared, a
team at the Nat i onal I nst i t ut e on Alcohol Abuse and Alcohol-
ism announced t hat it could find no si gni fi cant di f f er ence be-
Demon Rum I 99
tween alcoholics and nonalcoholics in the frequency of the Al
allele. In 1991 a group at Washington Uni v ersi t y in St. Louis
also failed to find any association. And, more recently, a team
at the Nat i onal Inst i t ut e of Mental Health, looking not at the
allele but directly at the D
2
gene, also f ound no di f f er ences
between alcoholics and nonalcoholics.*
None of these negative fi ndi ngs made the front page of any
newspaper, if they were reported at alla trend that is dis-
tressingly f ami l i ar to scientists in the field of genetics. The
initial announcement s of the "discoveries" of genes causing
manic-depression, schizophrenia, and excessive violence among
men with an extra Y chromosome all were t rumpet ed loudly
in popular media. In each case, however, follow-up st udi es
fai l ed to confirm the i ni t i al findings, and in some cases the
earlier report s were ret ract ed (Horgan 1993). But, as happened
in the case of the D
2
gene, none of these ret ract i ons and cor-
rections received anyt hi ng like the original coverage. This ki nd
of uneven reporting has no doubt contributed to the popular
assumption t hat genes are all-powerful and that alcoholism is
purely a genetic disease.
Nature
Does all of t hi s mean that the D
2
theory of alcoholism is dead?
Not at all. Numerous st udi es continue to suggest that some-
thing is going on with dopamine in the brains of some alco-
*Blum and Noble were members of t hat team, and they disagree with
their coauthors about the meani ng of the resul t s. They claim t hat the
mut at i on in the gene possessed by alcoholics lies not in the part of the
gene examinedthe so-called exon sequences, which det er mi ne the
structure of the dopamine receptorbut in some other section regul at i ng
the number of receptors made. To date, this hypothesized "intron mu-
tation" has yet to be i dent i f i ed.
100 / Buzz
holies, part i cul arl y those with the most serious type of alco-
hol i sm. Evidence f r om twin st udi es, the use of dopami ne-
boosting drugs, and ongoing genet i c screening studies indicate
t hat the Al allele is correlated to some ext ent with severe al-
coholism. By one estimate, def ect s in the D
2
gene may account
for about one-t hi rd of the overall i nf l uences on the prodigious
use of addictive subst ances ( U hl et al. 1993). Other, unknown
genes were est i mat ed in this st udy to aceount for another t hi r d,
and the l ast t hi rd was at t r i but ed to the env i r onment . As the
aut hor s of t he st udy noted, t hi s model would leave D
2
mut a-
t i ons as "one of the most pr omi nent single gene det er mi nant s
of suscept i bi l i t y to severe substance abusebut ot her genes
and the env i r onment , when combined, still play the largest
role."
That other genes, and ot her neur ot r ansmi t t er s, probably
have a hand in alcoholism is har dl y surpri si ng given the brai n' s
complexity. In f act , v ari at i ons i n several ot her neurot ransmi t t cr
systems have been t ent at i v el y l i nked to alcoholism. One can-
di dat e is serot oni n. The br ai ns of alcohol-preferring rat s, for
i nst ance, have been shown to have lower levels of serotonin
t han the br ai ns of other rats ( Mur phy et al. 1987). And several
ani mal and human exper i ment s have shown that: drugs boost-
ing s er ot oni n (such as Proz ac) tend to modestly reduce dr i nk-
ing ( Depar t ment of Health and Human Services 1993).
All t hese st udi es suggest a rel at i onshi p between i nt ernal se-
rot oni n l ev el s and the desi r e for alcohol. But none of the as-
soci at i ons or ef f ec t s observed arc so s i gni f i cant that serot oni n
is being viewed as the al cohol -rel at ed neur ot r ansmi t t cr . For in-
stance, one st udy of alcoholics given ant i depr essant s t hat boost
serot oni n levels f ound t hat t he drugs af f ect ed only about hal f
of the st udy subj ect s, and t hat in t hi s group t her e was only a
20 to 30 percent reduct i on in dr i nki ng ( Nar anj o et al. 1990).
Such r esul t s may be expl ai ned by the hypot hesi s that serotonin
Demon Rum I 101
is more responsible for regul at i ng mood than for controlling
alcohol cravings. In people s uf f er i ng an underl yi ng mood dis-
order, t hen, correcting serotonin levels could reduce t hei r need
for alcohol, while those whose disorder l i es elsewhere derive no
benefi t .
Other research suggests that defect s in the endorphi n system
of alcoholics may cont ri but e to the condi t i on. It has also been
suggested t hat genetically based di f f er ences in alcohol meta-
bolism may play a role. The bottom line is that the view of
alcoholism as a single, clear-cut disease is considerably weaker
and less substantiated t han most people t hi nk. It' s not that
genes have not hi ng to do with alcoholism. We've j ust seen
excellent evidence that they are involved. But the nat ur e of
that involvement and the degree to which it is mani fest ed in
a given alcoholic is still unknown.
For the moment, there is only the t heoryas yet un-
provedthat genes set a kind of background tone for alcohol
response. Many genes probably contribute to this tone. Some
may i nfl uence how the body metabolizes alcohol; others may
influence the sensi t i v i t y of GABA receptors; still otherssuch
as the i nf amous D
2
receptor genemay set the idle speed of
the internal pleasure-producing machinery. Vari at i ons in the
ways these genes work may leave some people more or less
sensitive to alcohol than others. But no mat t er how much ge-
netic v ar i at i on is at work, genes don' t f unct i on in a v acuum.
They are expressed in human beings who live t hei r lives in
diverse and complicated environments.
Nurture
Researchers probing the environmental side of the alcoholism
coin begin wi t h the obvious quest i on: Why do people dr i nk?
Instead of looking at the stage set by genes, these researchers
102 I Buzz
look at what people do. They look at the choices people make
in their dri nki ng and at their behavior toward alcohol in gen-
eral.
What they' ve f ound runs the gamut from the obvious to the
surprising.
Many researchers were surpri sed, for example, by the results
of st udi es on the rel at i on of dri nki ng to stress. As we've seen,
alcohol mimics the actions of the ant i anxi et y drug Valium,
which might suggest that those experiencing marital, eco-
nomic, or job-related st ress would feel an increased urge to use
alcohol. But st udy af t er st udy has found only small or negligible
correlations between stress levels and dri nki ng levels. Basically,
the amount people dr i nk has l i t t l e to do with how much stress
t hey' re under .
Anot her env i ronment al f act or long t hought to cont ri but e to
alcoholism is expectancy. Some people regard alcohol as a
"magic el i xi r " capable of enhanci ng social skills, sexual plea-
sure, confidence, strength, and aggressiveness. When such be-
l i ef s have been acquired prior to the development of drinking
problems, they have been associated with increased risk of al-
coholism. This i sn' t surpri si ng given what we l earned in the
previous chapter about how one's beliefs about the potency of
alcohol change not only the subj ect i v e experience of consum-
ing alcohol, but the physical responses as well.
One of the probable ways that many people acquire their
expectations about alcohol is through advertising. The vigorous
promotion of alcohol consumpt i on in i t sel f, however, appar-
ently does not encourage people to dr i nk. A bcfore-and-aft cr
st udy of towns banni ng beer, wine, and liquor adv ert i si ng
found no subsequent change in total alcohol consumption
(Smart 1988). And the l i f t i ng of advert i si ng restrictions in Sas-
kat chewan, Canada, had no overall ef f ect on alcohol consump-
Demon Rum I 103
tion, though drinking shi ft ed slightly from spirits to beer
(Makowsy and Whitehead 1991).
Another nongenetic component of alcoholism could be pa-
rental influence. Here, too, the evidence is surprising. Harburg
and colleagues (1982) found t hat teenagers were more likely to
reject than emulate parental behavior when parental dri nki ng
became ext reme.
Although they are less influential than most people think, the
kinds of environmental factors j ust mentioned probably interact
in complicated and unpredictable ways with an individual' s ge-
netically set biological makeup. Accidents, viruses, severe abuse,
emotional t rauma, and learning can all change the brain' s cir-
cuitry; purely environmental factors, in other words, can alter
the biological foundat i ons of behavior. And genetic factors
how we look, where our talents lie, and so f or t haf f ect the way
others treat us and the way we experience the world, and thus
may exert considerable influence on the behaviors we develop.
As one scientist put it, "Genes and environment loop out into
each other and feed back on each other in a complex way that
we have j ust begun to understand" (Mann 1994).
The Spice of Life
Human biological diversity is hardl y a new concept. In the un-
settled years during which Jul i us Caesar struggled to gain con-
trol of the Roman Republic, the poet Titus Lucretius Cams
wrote a remarkable di dact i c poem called De Rerum Natura (On
the Nat ure of Things). The work is a lavish ode to Lucret i us' s
philosophical hero, the Greek philosopher Epi curus, who,
among other things, anticipated by thousands of years such
modern ideas as the atomic theory of mat t er, the uni v ersal i t y
of physical laws, and the molecular basis of individual di f f er -
104 I Buzz
ences. "With the out ward di f f er enc e between the v ari ous types
of animal t hat take food, " Lucr et i us wrote, "there go corre-
sponding di f f er ences in the shapes of t hei r component atoms.
These in their t ur n entail di f f er ences in the chinks and chan-
nelsthe pores, as we call t hemi n all part s of the body.
Subst i t ut e DNA for "component atoms" and "ion channels"
for "chinks and channels, " and you have a good approxi mat i on
of t oday' s view of the wel l spri ngs of human i ndi v i dual i t y. It is
now clear that if we could see a person' s chromosomes as
clearly as we can see her face, we woul d perceive in t hose long,
spiral molecules the same degree of i ndi v i dual i t y.
If the acceptance of uni que faces and bodies is anci ent , how-
ever, the notion that brains are equally uni que has taken longer
to root. Perhaps it is a r ef l ex belief t hat despite out ward dif-
ferences, inside we are all "created equal. " But , of course,
brai ns are no more the same from person to person than are
fi ngerpri nt seach is an expression of a singular genetic heri-
tage. Each of us has a uni que number of neurons, neuronal
connections, levels of neurot ransmi t t ers, and sensi t i v i t i es in
our ion channel s, and t hus uni que responses to outside i nf l u-
ences such as alcohol. ' I ' he cur r ent struggles to pin down the
mechani sms underl yi ng both or di nar y intoxication and al-
coholism are driving this point home with a vengeance. The
number of ways people v ary in the details of t hei r neur al ar-
chitecture, and specifically in t hei r responses to alcohol, is
ast oundi ng.
Some people arc physiologically v ul nerabl e to the ravages of
alcoholism; others can t ake alcohol or leave it. Certain individ-
uals get sleepy on low doses of alcohol and revved up on high
dosesexactly the reverse of what most people experi ence. A
person might get hangovers on whi t e wi ne but not red, or re-
quire t wo hours r at her t ha n one t o recover ment al cl ari t y af t er
a single dr i nk. Such v ar i at i ons can make hash of at t empt s to
Demon Rum I 105
say anyt hi ng categorical about how people respond to alcohol.
Still, most of the ef f ect s discussed t hus far, from the molecular
to the behavioral, are t rue to some extent for all people. Un-
derstanding how alcohol usually works can provide a bench-
mark against which to measure one's own responses.
In our exploration of alcohol, we st art ed with the perspective
of a single moleculea pudgy, dog-shaped assembly of nine
atoms. By under st andi ng the size, shape, and chemical prop-
erties of ethanol, we saw why it so easily soaks into the body
and i nsi nuat es itself into the molecular machi nery underl yi ng
funct i ons as diverse as thought and ur i nat i on. Then we pulled
back a bit. We met other molecules as we followed a shot of
scotch down the throat. Proteins. Ion channels. Enzymes that
rip atoms off alcohol molecules. Underst andi ng something
about how these molecular machines work helps in under-
standing how alcohol itself works. Aft er t hat , we pulled back a
bit f ar t her , to the size of cells. We met neurons, the f unda-
mental uni t s of consciousness, and saw how they generate ac-
tion pot ent i al s, the "sparks" underl yi ng all human behavior.
Pulling back even f ar t her we looked at the body as a whole
at how moderate doses of alcohol can help the heart, modify
sexual response, or tweak the i mmune system. We also saw
how heavy or long-term dr i nki ng can rui n these bodily systems
and lead to impotence, enfeebl ement , pain, or death. We then
considered whole populations of dri nkers: those for whom al-
cohol is addi ct i ng and those for whom it is not. We looked at
how the nat ur e of one's genes is j ust the flip side of the nur t ur e
of one's envi ronment .
Now, ponderi ng the foundat i ons of our i ndi v i dual i t y, we are
back at the level of molecules. We've seen how our uni que
DNA gives rise to uni que brai ns, which in t urn give ri se to
uni que mi nds and personal i t i es. The complexity of the human
brain is the reason t hat alcohol is such a rich, complicated,
i 06 / Buzz
exasperat i ng s ubj ect . The molecule i t sel f is l aughabl y simple
as bori ng and st at i c as a pinball. But let a few t ri l l i on of those
pi nbal l s f al l i nt o the machi ner y of the mi ndi nt o the flashing,
dea f eni ng conf abul at i on that is a human bei ngand anyt hi ng
can happen. Anyt hi ng at all.
The World's Favorite Drug
Alcohol is scarce in the nat ur al world. Producing appreciable
quant i t i es demands somewhat laborious and delicate manipu-
lations of yeast. Caf f ei ne, in contrast, qui t e l i t eral l y grows on
trees. And bushes. And some ki nds of cactus. And some species
of lily and holly and camellia. In f act , at last count, more than
a hundred pl ant species produce caf f ei ne molecules in their
seeds, leaves, bark, or other st ruct ures, making for a t rul y re-
markable di st ri but i on (Viani 1993). Two other popular plant-
produced moleculesnicotine and morphineare roughly the
i 08 / Buzz
same size and complexity as caf f ei ne, but both arc produced
in only a single plant species: tobacco (Nicotiana tabacum) and
opium poppies (Papaver somniferum), respectively.
Since caf f ei ne- cont ai ni ng pl ant s grow almost everywhere in
the tropics, it is not sur pr i si ng t hat the i nhabi t ant s of those
regions long ago l earned ways to ext ract the st i mul at i ng drug
for t hei r own uses. In Afri ca, caf f ei ne was discovered in kola
nuts and in the seeds and leaves of the many species of coffee
t ree, two of which are grown commercially: Coffea arabica and
Coffea robusta. Arabica beans arc harder to grow, produce more
flavorful cof f ee, and cont ai n about half t he caf f ei ne of robusta
beans. In Chi na, c a f f e i n e was discovered in the leaves of tea
plants. And in South America, caf f ei ne was f ound in the leaves
of the mate plant (used to make a dr i nk of the same name) as
well as in the seeds or berries of several ot her pl ant s used to
make beverages no longer popul ar .
Tea and coffee have become the most popul ar dri nks on
eart h. Aside f r om plain water, more tea is consumed every day
by the world' s people t han any ot her single beverage (Graham
1984). Cof f ee is a close second, and because a t ypi cal cup of
cof f ee cont ai ns about twice as much caf f ei ne as a cup of tea,
cof f ee is act ual l y the single largest source of c a f f ei ne world-
wi de (Gilbert 1984). In the Uni t ed States, soda, not tea, is
the most popul ar beverage; per capita sof t - dr i nk consumpt i on
in 1993 was nearly 50 gal l ons. Cof f ee ranked second with 34
gal l ons consumed per person a year, and beer was t hi rd at
about 23 gallons ( Berry 1994; Nat i onal Cof f ee Association
1991).
The popul ar i t y of soda in the Uni t ed States hardl y means
t hat Ameri cans pr ef er caf f ei ne- f r ee beverages. Roughly 86 per-
cent of the 12.7 billion gallons of soda consumed in 1993 con-
t ai ned c a f f ei ne (Berry 1994). A good deal of t hi s caf f ei ne is
f ound in cola dr i nks such as Coca-Cola and Pepsi-Cola and
The Eyelids of Bodhidharma I 109
their many i mi t at ors. In fact , the word "cola" comes from
"kola," the name of the Af r i can tree that produces the caffei ne-
containing seeds from which a flavor ext ract is made. This kola
ext ract was one of the ingredients in the original recipe for
Coca-Cola, invented by Georgia pharmacist John Pemberton
in 1886. Pemberton' s brew also contained cocaine, derived
from the coca plant of South America, which is where the
"coca" in Coca-Cola" comes from. Af t er the addictive poten-
tial of cocaine was recognized around the turn of the cent ury,
the drug was el i mi nat ed from the recipe and replaced with
caf f ei ne. Today, both Coke and Pepsi contain about 45 milli-
grams of caf f ei ne per 12-ounce canroughly the same as a cup
of tea or half a cup of coffee.
But non-cola soft dri nks can contain significant amount s of
caf f ei ne as well. Mount ai n Dew, for instance, contains 54 mil-
ligrams of caf f ei ne per can. Mellow Yellow and Dr. Pepper
also contain hef t y doses. Int erest i ngl y, almost all the caf f ei ne
in these dri nks is purchased by soda manuf act ur er s from the
makers of decaf f ei nat ed coffees and teas, for which caf f ei ne is
a valuable by-product indeed.
In America, t herefore, soda consumption accounts for a sig-
ni fi cant percentage of total caf f ei ne ingestion. Despite the phe-
nomenal growth in an espresso-based cafe culture in many
large cities, coffee consumption overall has been declining
slowly over the past decade while soda consumption has risen
steadily. If current trends cont i nue, more Americans will get
their caf f ei ne buzz from soda by the t ur n of the cent ur y t han
from any other source.
A Brief History of Caffeine
The world' s fondness for caf f ei ne has been a long-standing love
a f f a i r . Although the discovery and use of caffei ne-cont ai ni ng
110 I Buzz
pl ant s predat es writing, v ari ous legends and myt hs about t he
discovery of eof f ee and tea have survived to t he present day.
The discovery of tea is at t ri but ed to the Chinese emperor
Shen Nung. The year is fixed as 2737 B.C. According to legend,
one evening the emperor was boiling water in an open kettle
over a carnpfi re built from the branches of a nearby shrub.
Some scorched leaves from these branches swirled up in the
col umn of hot air and fel l back into the water. Rather than
di scardi ng the cont ami nat ed water, the emperor t ast ed it and
was intrigued by the ast ri ngent taste and refreshi ng aroma. Fur-
ther exper i ment at i on with more leaves of the same tree con-
vinced Shcn Nung of the value of the plant as a health-giving
medi ci ne. Over the cent uri es, the use of tea expanded from its
i ni t i al role as a medicinal herb to t hat of a ubi qui t ous social
beverage.
The cust om of dri nki ng tea was brought from China to Japan
by Buddhi st pri est s around the year A. D. 600 (McCoy and Wal-
ker 1991). This expl ai ns why the legend of tea' s origin in Japan
is linked to Buddhi sm, and in part i cul ar to Bodhidharma, the
sage who f ounded the Zen branch of Buddhi sm. According to
the legend, Bodhi dharma fell asleep in the course of an ex-
t r emel y long medi t at i on. Disgusted with his own weakness, he
tore off his eyelids and flung t hem to the ground. Where the
eyel i ds fell, tea pl ant s sprang up, t hus providing other Buddhist
priests with a tool for ext endi ng the reach and power of t hei r
medi t at i on.
The connection between c a f f e i n e and religious devotion also
fi gures pr omi nent l y in one of the common legends about cof-
f ee. In t hi s myt h, a sharp-eyed Arabian goatherd named Kaldi
noticed hi s flock munchi ng the bright red, cherryl i ke f r ui t of a
shr ub nat i v e to nor t heast er n Af r i ca. Soon a f t er the goats ate
the berri es, t hey began prancing around with unus ual gusto.
Kaldi tried the berri es himself and was so refreshed and invig-
The Eyelids of Bodhidharma / I I I
orated t hat he danced along with his goats. This frolicsome
behavior was noticed by a drowsy monk who was passing by on
his way from Mecca. Impressed, the monk asked Kaldi the se-
cret of his energy. Kaldi showed him the berri es. The monk
was delighted to find t hat he could now pray longer and wi t h
more at t ent i on. He spread the word to his fellow monks, who
experi ment ed with other ways to consume the berries. Even-
tually, people f ound that roast i ng the seeds, gri ndi ng them up,
and soaking them in hot water produced a beverage that was
tasty and gave a great er "kick" than could be achieved by
merely chewing the caf f ei ne- cont ai ni ng f r ui t and seeds.
Nature's Pesti ci de
As wi t h the ar t s of f er ment at i on and di st i l l at i on, humans mas-
tered the cul t i v at i on, processing, and preparat i on of caf f ei ne-
containing beverages long before they knew what gave these
dri nks t hei r z ip. But whereas alcohol was isolated f r om wine
and beer by distillation in the Middle Ages, the active ingre-
dient in cof f ee and tea remai ned myst er i ous unt i l the nine-
teenth cent ur y.
The reason is t hat the separat i on of caf f ei ne f r om tea and
coffee is considerably more di f f i cul t than distillation. It re-
qui res several separate chemical steps and the use of a st r ong
solvent, such as hexane or chl oroform. It wasn' t unt i l 1820, at
the dawn of the modern era of organic chemi st ry, that caf f ei ne
was discovered. The word "caffei ne" comes from cafe, the
French word for "coffee. " The word "coffee," in t ur n, was
coined by the great nat ural i st Carolus Linnaeus. l i e gave the
name Coffea to the genus of tropical shrubs now called cof f ee
trees. Linnaeus created his word as a Lat i ni z at i on of t hr ee of
the many di f f er ent words then used to describe cof f ee: "caova,"
"cova," and "kahwah."
I 12 / Buzz
By 1865 c a f f e i n e had been isolated not only from cof f ee, but
f r om t ea, mat e, kola n u t s , and v ar i ous South Ameri can pl ant s
used to make abori gi nal dr i nks . At a phar maceut i cal meet i ng
that year a Professor Bentley noted what many have com-
ment ed on si nce. "It is most remarkabl e, " he said, "that all the
most i mpo r t a n t unf e r me nt e d beverages i n use i n di f f er ent part s
of the globe shoul d be prepared from subst ances containing
the same or a closely al l i ed al kal oi d" ( Ki hl man 1977). It isn' t
clear whet her Bent l ey was more impressed by t he prodi gi ous
huma n t hi r s t for st i mul ant s or by t he unus ual l y wide di st ri -
but i on of c a f f e i ne among pl ant s, but both f act s are, indeed,
r emar kabl e. The al kal oi d f ami l y r ef er r ed to by Bent l ey is a large
and general l y poisonous group of ni t rogen-cont ai ni ng com-
pounds t hat i ncl udes s t r ychni ne, ni cot i ne, mor phi ne, rnesca-
line, and emet i ne; the l ast of these is the deadly i ngredi ent in
poison heml ock, the herb used by the anci ent At heni ans to
execut e the philosopher Socrates.
As we' ve seen, most al kal oi ds are speci fi c to a par t i cul ar plant
species. Mow is it, t hen, t hat c a f f ei ne is f ound in pl ant s as
unr el at ed as cert ai n species of l i l i es and cacti? Why do more
t han a hundr ed species go to the t roubl e of ma nuf a c t ur i ng this
one psychoactive molecule? One t heor y holds that c a f f ei ne
hel ps pl ant s ward off at t ack f r om insects and ani mal s; c a f f ei ne
may make seeds, leaves, and other pl ant part s t ast e bi t t er , t hus
di scouragi ng consumpt i on by predat ors. It is also possible t hat
caf f ei ne acts di r ect l y on ani mal ner v ous systems to create un-
comf or t abl eor downri ght l et hal ef f ect s . For both these rea-
sons, caf f ei ne- cont ai ni ng pl ant s may have a di st i nct adapt i v e
edge over ot her pl ant s , which would logically l ead to their pro-
l i f e r a t i on over other species.
The caf f ei ne- as - dcf cns e theory is support ed by research on
c a f f e i ne ' s pest i ci dal propert i es. In the mid-1980s, James Na-
t hans on of Harv ard Medi cal School f ound t hat when prepara-
'I'he I'lyelids of Bodhidharma I I 13
tions of tea leaves or cof f ee beans were fed to l arv ae of young
tobacco hornworms, the insects ate poorly and s uf f er ed t rem-
ors, hyper act i v i t y, and s t unt ed growth ( Nat hanson 1984). At
slightly higher concent r at i ons, the l arv ae were killed wi t hi n
t went y- f our hours. The same was t r ue for mealworm l arv ae,
a species of but t er f l y, mosqui t o l ar v ae, and milkweed-bug
nymphs. The ef f ect s were so pronounced t hat Nat hanson sug-
gested t hat c a f f e i ne be considered a viable nat ur al al t er nat i v e
to stronger and more ecologically damaging pest i ci des.
Similar ki nds of chemical sel f - def ense systems have been dis-
covered in many other pl ant s, so the role of caf f ei ne as a ki nd
of na t ur a l nerve poison makes a good deal of sense. In f act ,
the same ev ol ut i onar y pri nci pl es may expl ai n the existence of
ni cot i ne, morphi ne, cocaine, t et rahydrocannabmol (t he active
i ngredi ent in ma r i j u a n a ) , and ot her pl ant -deri ved psychoactivc
drugs. These compounds probably exi st because t hey di s r upt
the nerv ous systems of insects and ani mal s, t hus conf er r i ng an
adapt i v e adv ant age t o pl ant s t hat manuf act ur e t hem.
This surv i v al adv ant age is probably not, however, the ent i re
expl anat i on for caf f ei ne' s presence in so many ki nds of plants.
If it were, the growers of tea and cof f ee would have a much
easier job. But despite the presence of caffei ne-l aced sap in the
t i ssues of tea pl ant s and coffee trees, many insects find both
crops emi nent l y palatable, forci ng growers to rely heavi l y on
pesticides. (In recent year s, some growers have r et ur ned to or-
ganic approaches with some success, suggesting that the prac-
tice of growing dense monocul t ures of such crops as coffee and
tea may cont r i but e to t hei r v ul ner abi l i t y to insect at t ack. )
The presence of caf f ei ne in tea and cof f ee pl ant s has, of
course, provided these species with a ser endi pi t ous adv ant age:
because c a f f e i ne is so highly prized by humans, the pl ant s from
which it is derived have been protected and di st r i but ed to an
ext ent impossible under nat ur al condi t i ons.
I 14 / Buzz
The point is that caf f ei ne' s pesticidal properties, though real,
are probably not the whole reason this compound appears in
so many plants.
Caffeine: The Molecule
Anot her expl anat i on for caf f ei ne' s wide di st ri but i on in the
plant kingdom is that it' s manuf act ur ed f r om a very common
raw i ngredi ent . The st art i ng point for caf f ei ne production is a
ubiquitous molecule called xant hi ne:
Xanthine
Xant hi ne is f ound in both plants and animals. It is a raw ma-
terial used const ant l y in the creation and maintenance of DNA.
Xanthine isn' t usually f ound in great quant i t i es in animal bod-
ies because it is rapidly recycled into other molecules by en-
zymes. In humans, for instance, excess xant hi ne is usual l y
converted into uri c acid, whichas the name suggestsis ex-
creted in uri ne. In plants that make caf f ei ne, however, xant hi ne
is shunt ed into an enzymatic assembly line that methodically
attaches common molecular uni t s called methyl groups. A
methyl group is a carbon with three hydrogen atoms attached.
When a methyl group is attached to xant hi ne, a methylxan-
thine is f or med. When a second is added, a di met hyl xant hi ne
is formed. And when a third is attached, a fnmet hyl xant hi ne
The Eyelids of Bodhidharma / M S
is created. Caf f ei ne is a part i cul ar kind of t r i met hyl xant hi ne.
Here is its molecular picture:
Caffeine
As you can see, caf f ei ne is a globular molecule with a knobby
sur f ace. One consequence of this l umpi ness is that caf f ei ne is
far more selective in its actions t han is alcohol. Like a compli-
cated key that can work only in an equally complicated lock,
the uni que bumps and grooves of caf f ei ne mean that it i nt er-
acts only with molecules that happen to mi rror those shapes.
This makes caffei ne very picky in its act i onsj ust the opposite
of alcohol's tendency to af f ect practically every molecular sys-
tem it touches. Caf f ei ne, in f act , leaves the vast maj or i t y of
the body's molecules alone. But it binds very strongly indeed
to the select handf ul that happen to have the correct shape.
In practical terms, this means that it takes much less c a f f ei ne
than alcohol to achieve a desired physical or mental ef f ect .
Unlike the st andard alcoholic dri nk r ef er r ed to earl i er, there
is no of f i ci al l y sanctioned "standard drink" of caf f ei ne. The
most commonly used uni t is the 100 milligrams of caf f ei ne
found in an average (8-ounce) cup of regular cof f ee. Just as a
shot of whiskey, a glass of wine, and a can of beer cont ai n about
a hal f-ounce of alcohol, an average cup of cof f ee, two cups of
black tea, and two 12-ounce cans of caf f ei nat ed soda contain
roughly 100 milligrams of caf f ei ne. Notice that we' re talking
I 16 I Buzz
milligrams here. A mi l l i gr am is a t housandt h of a gram
r oughl y the weight of a single grain of salt. In cont rast , the
s t andar d dose of alcohol is measur ed in l i qui d ounces. The hal f -
ounce of alcohol in a s t a nda r d dr i nk is equi v al ent to 14,200
mi l l i grams. The st andar d dose of alcohol is t hus roughl y 142
t i mes l ar ger t han t he st andar d dose of c a f f ei ne.
Act ual l y, t hi s compari son says more about t he rel at i v e i m-
potence of alcohol t han t he st rengt h of c a f f ei ne. Caf f ei ne' s
potency i s s i mi l a r t o t hat of ot her f a mi l i a r dr ugs like aspi ri n,
whi ch is t ypi cal l y admi ni s t er ed in doses of bet ween 50 and 500
mi l l i gr ams. An exampl e of a truly potent dr ug is lysergic acid
di ct hyl ami de, or LSD. A t ypi cal "hi t " of LSD is a mere t ent h
of a mi l l i gr am. Picture cutting a single grain of salt into ten
pieces and us i ng j u s t one of those pieces. That' s how l i t t l e LSD
i t t a kes to produce a neurol ogi cal i mpact far more dr amat i c
t han t hat caused by a cup of cof f ee.
As d i f f e r e n t as they are in potency, however, caf f ei ne and
alcohol molecules have an i mpor t ant si mi l ar i t y: both dissolve
easi l y in ei t her wat er or f a t . Just l i ke alcohol, caf f ei ne molecules
are qui ckl y absorbed in the small i nt est i ne (and, to a lesser
e xt e nt , i n t he stomach i t s e l f ) . Ca f f ei ne easi l y crosses cell mem-
branes, and is r api dl y di st ri but ed to all of the body' s t i ssues. It
also d i f f u s e s i nt o bodi l y l i qui ds such as saliva, semen, breast
mi l k, and amni ot i c f l ui d. Unl i ke alcohol, however, hardl y any
caf f ei ne is el i mi nat ed in the breat h or ur i ne: it simply ci rcul at es
unt i l it is destroyed by enz ymes in the l i v er.
Caffeine's Cousi ns
Caf f ei ne isn' t the only mct hyl xant hi ne consumed by hu-
mans i t ' s j us t t he most f amous. Cof f ee and t ea also contain
v er y small amount s of a r net hyl xant hi ne called theophylline.
The Eyelids of Bodhidharma I I 17
Cacao product s ( t hat is, chocolate in all its guises) contain yet
anot her met hyl xant hi ne called theobromine:
Theophylline Theobromine
As you can see, both theophylline and theobromine are di-
met hyl xant hi nes. Unl i ke caf f ei ne, these molecules have only
two methyl groups attached to t hei r xant hi ne skeleton. You can
also see that these compounds are extremely similar to each
other: they contain exactly the same number and kinds of at-
oms. They di f f e r only in the position of one of the methyl
groups. In a beaut i f ul example of how shape is everything in
chemistry, this seemingly trivial di f f er ence produces striking
di f f er ences in the way the two substances af f ect the brain. The-
ophylline is roughly as potent as caf f ei ne; theobromine is seven
times weaker than ei t her.
Theophylline is perhaps best known for its medicinal quali-
ties. Because it very effect i v el y relaxes the bronchial passage-
ways, it is of t en the drug of choice for treating asthma and
other breat hi ng di f f i cul t i es, such as congestion caused by pet
allergies. Indeed, allergy s uf f er er s who are unaware of theo-
phylline' s similarity to caf f ei ne may find themselves wide awake
at night or unusual l y j i t t er y if they take t hei r medication and
also dri nk their usual cof f ee, tea, or cola.
Caf f ei ne opens up bronchial passages alsothough less dra-
matically than theophyllineand this part i cul ar at t r i but e is
I 18 / Buzz
part of the story behind one of the most f amous coffee slogans
of all t i me.
Theodore Roosevelt was prone to ast hma at t acks when he
was a boy. His doctor recommended small doses of cof f ee to
ar r est these at t acks, which st art ed Roosevelt on a habit t hat
grew over time into a l egendary appet i t e (Siegel 1989). His
cof f ee mugs were said to more closely resemble vats than cups.
In 1907 cof f ee mer chant Joel Clark sought to t ake advant age
of t hi s prodigious t hi r s t . Cl ark had set up a booth to di spl ay
his wares at a count y f a i r to which Roosevelt was paying a vi si t .
When Roosevelt wal ked by his boot h, Clark t hr ust a cup of his
cof f ee at hi ma cup the presi dent prompt l y drained in a gulp.
Setting down the empt y cup, Roosevelt t urned to the people
ar ound him and declared the cof f ee "good to the last drop,"
t hus giving Clark and his brand of cof f eeMaxwel l House
a slogan that lives to t hi s day.
Caf f ei ne' s ot her cousin, the rel at i v el y weak theobromine, is
consumed by count l ess mi l l i ons every clay. Most people know
t hat chocolate cont ai ns a smal l amount of caf f ei ner oughl y
20 mi l l i gr ams in a 1-ounce por t i on. That' s not muchonl y
one- f i f t h the amount in an average cup of cof f ee. But most
people don't realiz e t hat chocol at e also cont ai ns theobromine.
In f act , t heobromi ne is seven times more abundant than caf-
fei ne in chocol at eabout 130 mi l l i grams in a 1-ouncc piece.
This abundance neat l y compensat es for theobromine' s lack of
raw pharmacol ogi cal punch. Basi cal l y, when theobromine' s in-
fluence is added to c a f f e i ne ' s , a 1-ounce piece of chocolate can
be said to have the s t i mul a t i ng power of roughly 40 milligrams
of caf f ei ne, about the same as t hat f ound in a cup of tea (Gil-
bert 1992).
If it were only ast hmat i cs who needed to appreci at e theoph-
yl l i ne, or "chocoholics" who needed to t hi nk about theobro-
mi ne, these two met hyl xant hi ncs would be mere curi osi t i es to
The Eyelids of Bodhidharma I I 19
most other people. But , in f act , these two substances play an
i mport ant role in the lives of everyone who dri nks coffee, tea,
colas, or any other caf f ei ne- cont ai ni ng beverage.
The Long Buzz
Regardless of your f ondness for a steaming mug of joe, a f r a-
grant cup of tea, or an ice-cold Coke, your body responds to
the caf f ei ne in these beverages as though you had j us t swal-
lowed poison. As with alcohol, liver enzymes are marshaled to
at t ack the molecules and disable them as qui ckl y as possible.
The human liver disposes of caf f ei ne by undoing the steps that
led to its f or mat i on in plants: methyl groups are plucked off
one at a t i me. This is an i mport ant poi nt : depending on which
methyl group is removed, caf f ei ne is t r ansf or med into theoph-
ylline, theobromine, or another di met hyl xant hi ne called par-
axant hi ne.
As we j us t saw, theophyllme is roughly as potent as caf f ei ne,
so when theophylline resul t s f r om the first stage of caf f ei ne
metabolism the arousing ef f ect s of the original caf f ei ne remai n
unchanged. Theobromine is only one-seventh as potent as caf-
fei ne, so the conversion of caf f ei ne to this di met hyl xant hi ne
does represent progress. But 70 percent of a given dose of caf-
fei ne is converted to par axant hi ne, which is act ual l y slightly
more potent t han caf f ei ne. This means that the "buzz" you get
from a cup of cof f ee lias as much to do with the breakdown
products of caf f ei ne as with the caf f ei ne i t sel f . Exactly how
paraxant hi ne a f f ec t s the brain is not clear, though it seems to
mimic the actions of caf f ei ne due to the si mi l ari t y of its
methyl-group conf i gur at i on. (We'll delve into these actions in
the next chapt er. )
In the second step of caf f ei ne metabolism in humans, an-
other methyl group is removed, producing a met hyl xant hi ne,
120 / Buzz
which has no s t i mul at i ng ef f ect s . From t her e, t he l ast met hyl
group i s remov ed, yi el di ng pl ai n- ol d xant hi ne, which ei t her i s
el i mi nat ed in ur i ne or is r eused. The pharmacol ogi cal act i v i t y
of t hcophyl l me, t heobromi ne, and par axant hi nc is part of the
reason it t akes a r el a t i v el y l ong t i me for a cof f ee buz z to wear
o f f . Not only must t he c a f f e i n e be el i mi nat ed, but t he break-
down pr oduct s have to be el i mi na t ed as well. The time it t akes
for a dose of a dr ug to wear off is measur ed by a v al ue cal l ed
a hal f - l i f e. That ' s the t i me it takes for half of a dose to be
el i mi nat ed. The hal f - l i f e of caf f ei ne averages bet ween fi ve and
six hours, which is far slower t han the rat e at which we elim-
i nat e alcohol. As l ei sur el y as c a f f e i ne ' s hal f - l i f e is, however, it
can be even longer for certain people.
Women t aki ng oral cont racept i v es r equi r e about twice the
normal t i me to el i mi nat e c a f f ei ne (Yesair 1984). For such
women, the st i mul at i on f r om a single cup of cof f ee might last
all day. A si mi l ar, t hough less dr amat i c i ncrease in caf f ei ne' s
h a l f - l i f e has been reported for women dur i ng the l ut eal phase
of the mens t r ual cyclethe t i me between ovulation and the
begi nni ng of mens t r uat i on. In one st udy, caf f ei ne el i mi nat i on
took about 25 percent longer dur i ng t hi s t i me, r esul t i ng in an
average h a l f - l i f e of 6.8 hour s (Arnaud 1993). And in i nf ant s ,
the hal f - l i f e of c a f f e i ne i s radi cal l y ext ended because t hei r l i vers
have not yet developed the enz ymes needed to break down
c a f f e i n e . A f ul l - t er m newborn r equi r es eighty hour s t o meta-
bol i z e hal f a dose of c a f f e i ne (Snel 1993). As i nf a nt s grow, t hei r
abi l i t y to process caf f ei ne al so grows. By the time a baby is
between t hree and five mont hs old, a dose of caf f ei ne will have
an average h a l f - l i f e of 14.4 hours. And by about six mont hs,
i nf a nt s have es s ent i a l l y t he same abi l i t y t o process c a f f e i n e as
a dul t s . Al t hough s t udi es have f ai l ed t o fi nd any adverse con-
sequences on i n f a n t s f r om t he c a f f e i ne consumpt i on of nur s i ng
mot hers, the ext r emel y l ong hal f - l i v es in young babies is one
The Eyelids of Bodhidharma I 12!
reason t hat many doctors advise breast -feedi ng mot hers to
avoid caf f ei ne al t oget her. (This applies to expect ant mot hers
as well, as we'll see l at er . )
But anot her segment of the popul at i on experiences exactly
the opposite ef f ect as women on oral cont racept i v es. By a st i l l
i mper f ect l y understood mechani sm, cigarette smoking "revs
up" the l i v er' s caf f ei ne- dest r oyi ng enz ymat i c machi nery (Be-
nowi t z et al. 1989). As a resul t , the hal f - l i f e of caf f ei ne among
smokers is reduced to an average of t hr ee hours (Parsons and
Neims 1978). This double-speed el i mi nat i on of caf f ei ne may
expl ai n the l ong-st andi ng observation that smokers dr i nk more
coffee t han nonsmokcrs. Smokers may simply be adj us t i ng
their c a f f ei ne i nt ake to mai nt ai n the same degree of stimula-
tion achieved by nonsmoker s.
I nt er est i ngl y, it is appar ent l y not t he nicotine in cigarette
smoke t hat i nduces l i v er enz ymes to work more ef f i ci ent l y. Cig-
aret t e smoke cont ai ns hundreds of other volatile, reactive com-
pounds, and it is appar ent l y a f ami l y of such compounds cal l ed
polycyelie aromat i c hydr ocar bons that t ri ggers the increased en-
z ymat i c act i v i t y.
The i mpact of smoking on caf f ei ne clearance is i mpor t ant
for those who qui t smoking. In one st udy, bl ood-caffei ne levels
j umped an average of 250 percent a few days af t er the subj ect s
had qui t smoki ngev en t hough they di dn' t change t hei r cof-
fee- or t ea-dri nki ng habi t s, ' [' his added caf f ei ne j ol t could easi l y
exacerbate t he anxi et y, i nsomni a, i r r i t abi l i t y, and ot her un-
pleasant sympt oms of nicotine wi t hdrawal experi enced by qui t -
ters.
Smokers who dr i nk coffee and other caf f ei ne- cont ai ni ng
dr i nks are juggl i ng t he pharmacol ogi cal ef f ec t s of t wo f a i r l y
powerful al kal oi ds: ni cot i ne and caf f ei ne. This juggl i ng is
mostly unconscious: they aut omat i cal l y a dj us t t hei r consump-
tion of both drugs to mai nt ai n a desired physical or me nt a l
122 I Buzz
st at e. But as many people know f r om experi ence, t hi s j uggl i ng
is t ri cky. Not only do variables such as food i nt ake and sleep
al t er the body' s response to both subst ances, but i nt er act i ons
such as the one j us t ment i oned between cigarette smoke and
c a f f ei ne met abol i sm can produce ef f ect s t hat can leave a user
grappl i ng with physi cal react i ons t hat seem out of proport i on
to the amount of a dr ug consumed. Under st andi ng the nat ur e
of c a f f ei ne and how it behav es in the body can i nf or m the sel f-
regul at i on e f f o r t s of smokers and nonsmokers al i ke.
Now t hat we know somet hi ng about caf f ei ne as a molecule,
we' re ready to t ake a look at what happens when those mole-
cul es hi t an unsuspect i ng br ai n.
The Taste of Pi tch
When cof f ee was i nt r oduced to Kur ope by ent er pr i s i ng mer-
chants in the middle of the seventeenth cent ury, it was re-
garded as an exotic Arabian curi osi t y at best, a r epul si v e excuse
for a beverage at wor st . Three hundr ed years ago, many people
coul dn' t imagine consumi ng a hot, bi t t er , black dr i nk. It re-
minded t hem of hot pi t ch, whi ch was used as a medieval
weapon and i ns t r ument of t or t ur e (Schivelbusch 1992). But
t hi s i ni t i al aes t het i c resi st ance t o cof f ee qui ckl y evaporat ed as
the pharmacological powers of the dr i nk became widely appre-
124 / Buzz
ei at ed. Wi t h i n a decade of i t s nea r l y si mul t aneous i nt r oduct i on
t o ur ba n s eapor t s such as London, Amst er dam, Veni ce, and
Mar sei l l es, cof f ee dr i nki ng was commonpl ace. Cof f eehouses
and cafes opened up by t he t housands, and comment s about
coffee' s ef f ect on ( ' he mi nd began a ppea r i ng i n t he pr ess. "Tis
f ound al r eady, " wr ot e J a me s Howcll i n 1660, "that t h i s coffee
d r i n k h a t h caused a great er sobri et y among t he Nat i ons.
Wher eas f or mer l y appr ent i ces and cl erks used t o t ake t hei r
morni ng' s dr aught " of al e, beer, or wi ne, whi c h by t he di z z i ness
t hey cause i n t he br a i n made ma ny u n f i t f or bus i nes s , t hey
now pl ay t he good-fel l ows i n t h i s wakef ul and civil dr i nk. "
The c a f f e i n e i n cof f ee, tea, and chocol at eal l of which were
i nt roduced t o Kur opc at a bout the same t i meneat l y dove-
t ai l ed wi t h t he i deal s and v al ues of Enl i ght enment society.
"Cof f ee f unc t i oned as a hi s t or i cal l y si gni f i cant drug, " wr i t es
Wol f gang Scbi v el bns ch ( 1992) . "It spread t hrough t he body
and achieved chemi cal l y and phar macol ogi cal l y what r at i onal -
i sm and t he Pr ot est ant et hi c sought t o f u l f i l l s pi r i t u a l l y and
i deol ogi cal l y. Wi t h cof f ee t he pr i nc i pl e of r a t i ona l i t y ent ered
huma n physi ol ogy. "
In s hor t , people i ns t a nt l y recogni z ed cof f ee for what it was:
a pot ent s t i mul a nt t hat i nduced a ment al and physical energy
t ha t was bot h pl eas ur abl e i n i t s own r i ght and emi nent l y us e f ul .
Cof f ee' s a b i l i t y to f ost er i ndust r i ousness was seized on by em-
pl oyer s l ooki ng to boost pr oduct i v i t y and cl ergy l ooki ng to re-
duce al cohol cons umpt i on among their flocks. An anonymous
poem, publ i shed in 1674 dur i ng the early year s of cof f ee' s pop-
ul a r i t y, is t ypi c a l of the ki nd of good press cof f ee of t en received
( Scl nv el busch 1992):
When the sweet poison of the t r eacher ous grape
Had acted on the worl d a general rape;
A Quicker Genius I 125
Drowning our Reason and our souls
In such deep seas of large o' erfl owi ng bowls,
When foggy Ale, levying up mighty t r ai ns
Of muddy v apours, had besieg' d our brai ns,
Then Heav en In Pity
First sent amongst us this Al l - heal i ng berry.
Cof f ee ar r i v es, t hat grave and wholesome Liquor,
That heals the stomach, makes the geni us qui cker,
Relieves the memory, revives the sad,
And cheers the Spi ri t s, wi t hout maki ng mad.
Coffee' s popularity has thus always been intimately tied to
its ability to powerful l y alter the funct i oni ng of the human
brainto make "the genius qui cker. " As we saw in the previ-
ous chapter, the compound responsi bl e for t hi s "quickening"
was isolated from cof f ee in 1820. But only in the past several
years have neur osci ent i st s made much progress in i l l umi nat i ng
how caf f ei ne revs up the brain.
Perchance, to Dream
The mechanics of caf f ei ne came to be understood duri ng
i nvest i gat i ons of one of its most obvious and well-known prop-
er t i es: the ability to t emporari l y bani sh sleep. The ni net eent h-
cent ur y homeopath Samuel Hahncmann recognized t hi s
cl ear l y when he wrote, in 1803, t hat for cof f ee dr i nker s "sleep-
iness v ani shes and an ar t i f i ci al spri ght l i ness, a wakef ul nes s
wrested f r om Na t ur e takes i t s place."
Nobody at that t i me had the slightest idea about how caf -
f ei ne worked, but , i nt erest i ngl y enough, they di d hav e what has
t ur ned out to be a good grasp of t he f undament al process un-
126 / Buzz
deri v i ng wakef ul nes s and sleep. It had long been suspected t hat
people get drowsy because of a bui l dup of some sort of sleep-
i nduci ng chemical produced dur i ng the day as the brai n
"worked." This was the r at her crude idea behind a pioneering
exper i ment conduct ed by two French sci ent i st s in t he earl y
years of the t went i et h cent ur y. Rene Legendre and Henri Pi-
eron worked with pai r s of clogs. One dog was allowed to sleep
normal l y, while the other was kept awake for extended periods
of t i me. (How the exper i ment er s managed to keep t hei r hapless
subj ect s f r om f al l i ng asleep isn' t rel at ed. ) They then ext ract ed
f r om the sleep-deprived dogs a small amount of the fluid that
cont i nuousl y bat hes both the brai n and the spinal cord. When
t hi s fluid was i nj ect ed into the brains of the well-rested dogs,
t hey pr ompt l y fell into a long slumber, proving that a sleep-
i nduci ng subst ance is, indeed, present in the brai ns of tired
ani mal s.
U nf or t una t e l y for Legendre and Pieron, the tools then avai l -
able for exami ni ng cer ebr ospi nal fl ui d for mi nut e amount s of
organic compounds were wholly i nadequat e to the t ask of iden-
t i f yi ng t he myst eri ous subst ance. The exper i ment er s had f ound
an i mpor t ant clue about the chemical under pi nni ngs of sleep,
but t hey never came any closer t o under s t andi ng i t s t r ue na-
t ure. Since t hat time, of course, sleep has been ext ensi vel y
probed, in both ani mal s and humans. Thanks to v ol unt eer s
who collectively have spent t housands of nights sleeping in
l aborat ori es with t hei r heads wired l i ke Chri st mas trees, we now
know t hat sleep is far more complex than a simple, f eat ur el ess
slumber. Brain-wave pat t er ns change dramat i cal l y t hroughout
the ni ght as we pass t hrough f our maj or phases of sleep; several
di st i nct br ai n st r uct ur es cooperate t o i nduce and mai nt ai n
sleep; and not one, but several subst ances have been put f or t h
as candi dat es for the sleep-inducing chemical sought by the
A Quicker Genius I 127
French researchers at the t ur n of the century. Recent st udy of
one of these substances, called adenosine, has revealed much
about how caffei ne works.
For a long time, adenosine wasn' t given much at t ent i on. It
was known pri mari l y for its role in the t r ans f er of energy in
cells. As cells use energy, adenosine is produced as a by-
product. The harder a cell works, the more adenosine is cre-
ated. The excess adenosine is pumped outside the cell. But
recent work has revealed t hat adenosine is much more t han
mere cellular exhaust . In yet another example of nat ure' s par-
simonious ways, the adenosine produced by working cells is
used as a signaling molecule in an elegant, sel f-regul at i ng con-
trol syst em. How t hi s control system works to regulate sleep
and wakef ul ness was described by Robert Greene and his col-
leagues at the Har v ar d Medical School (Rai nni e et al. 1994).
Duri ng the day, neurons fire frequent l y as we go about our
daily business. As t hey fire adenosine is produced and ends up
floating around in the immediate v i ci ni t y of the neuron. Em-
bedded in the membranes of neurons (and many other types
of cells) are receptors designed specifically for adenosine.
When adenosine latches onto one particular ki nd of adenosine
receptor, a chemical chain reaction is triggered inside the cell
t hat almost i mmedi at el y opens ion channels in the membrane.
The openi ng of these channels either directly inhibits a neuron
from firing or reduces the amount of neur ot r ansmi t t er released
into the synapse. In ci t her case, the net ef f ect is to dampen
activity of both the neurons producing adenosine and the neu-
rons in the immediate vicinity. Adenosine t hus f unct i ons as a
ki nd of t hermost at : it keeps neuronal act i v i t y within safe limits.
If neurons fire excessively, adenosine builds up, which slows
firing rat es. Less firing, of course, means less adenosine being
produced; hence adenosine i nhi bi t s its own product i on. If ac-
S 28 / Buzz
t i v i t y drops off too much, adenosinc levels fal l as well, rel easi ng
t he "brake" on f i r i ng and allowing ncur onal act i v i t y t o rise
agai n.
This sel f - r egul at i ng adcnosi ne "thermostat" is typically very
localiz ed: only the cell produci ng t he adcnosi ne and a few
nei ghbors are involved. But r ecent l y it has been l earned t hat
aclenosine pl ays a cruci al role in setting the overall arousal level
of the br ai n. ' I' wo t humbt ack- si z e patches of neur ons located
on the brai n stem are par t i cul ar l y loaded with adenosinc re-
cept ors. The neur ons in t hese two areas fan out and touch
nearly every ot her part of the br ai n. They arc t hus except i onal l y
power f ul s t i mul a t e t hem, and the act i v i t y of the entire brain
increases; dampen t hei r act i v i t y, and the ent i re brai n "goes to
sleep."
As we use our br ai ns whi l e we are awake, adenosinc builds
up in these areas and a "brake" is applied with ev er-i ncreasi ng
force, gradual l y qui et i ng act i v i t y all over the br ai n. We become
drowsy and feel a keen urge to sleep. Once asleep, the adcno-
sine out si de t he cells is reabsorbed and recycled for use in en-
ergy pr oduct i on the next day. As adenosi ne levels drop, the
"brake" is released and we wake up.
Blocking the Brake
Adenosi ne resembles anot her of the br ai n' s pot ent "brakes":
GABA, which, as we saw earl i er, is one of the neurochemical
substances used by the brai n to of f s et and balance equally pow-
er f ul "accelerator" ncur ot r ans mi t t cr s such as gl ut amat e. The
br ai n t hus resembl es a car with sev eral brake pedal s and several
accel erat ors, i nt er f er e wi t h any one of these pedals, and you' ll
a f f ec t t he speed and action of the ear.
Caf f ei ne works by get t i ng in the way of the adenosi nc brake.
A Quicker Genius I 129
The reason becomes clear when the two molecules are com-
pared:
Caffeine Adenosi ne
Although at first glance these molecules may look dissimilar,
close inspection reveals that each is built on an identical dou-
ble-ring of carbon and nitrogen atoms. The primary di fference
between the two molecules is the addition of a type of sugar
on the lower l eft side of the adenosine molecule. Otherwise,
the two molecules are really quite similar indeedto the point
that caffei ne easily masquerades as adenosine in the brain. I n
fact, because it's slightly smaller than adenosine, caffei ne fits
more snugly into adenosine receptors than does adenosine it-
sel f. This tighter fit enables caffei ne to plug the receptor, thus
preventing adenosine from binding.
Despite this aggressive attraction to the adenosine receptor,
caffei ne doesn't actually make a perfect fit. And this makes all
the di fference. I f caffei ne mimicked adenosine exactly it would
be a depressant, not a sti mul ant: it would simply exacerbate
and extend adenosine's natural inhibition in the brai n. But
when caffei ne binds to the adenosine receptor, the resulting
shi ft in the shape of the receptor molecule is slightly di fferent
from the warping that occurs when adenosine itself binds. As
a result, the chemical chain reaction normally initiated by
i 30 I Buzz
adenosi ne isn't triggered by caffeine. Caf f ei ne is an i mpot ent
impostor: i t bi nds wi t h gusto, but f a i l s t o l aunch the all-
i mport ant qui et i ng message del i v ered by adenosi ne. Dr i nki ng
c a f f e i ne is t hus l i ke put t i ng a block of wood under one of the
br ai n' s pr i mar y brake pedal s. Ca f f e i n e i s an indirect s t i mul ant :
br ai n act i v i t y speeds up because it can' t slow down. By i t sel f ,
t hen, c a f f ei ne can' t st i mul at e a nyt hi ng. It can onl y clear t he
way for the brai n' s own s t i mul ant s ncur ot r ans mi t t cr s such as
gl ut ani at e, clopamme, and t he endor phi nst o do t hei r j ob.
You can, t her ef or e, get wi red only to the ext ent t hat your nat -
ur al exci t at ory ne ur ot r a ns mi t t e r s suppor t i t .
Among other things, t hi s expl ai ns why i t ' s al most i mpossi bl e
to overdose on caf f ei ne. Even if every adenosi ne receptor in
your br a i n were bl ocked by c a f f e i ne , you could still f unct i on.
You would cer t ai nl y feel s t i mul at ed, since one of your brai n' s
mai n "brakes" would be di sabl ed. But other brakes, such as
GABA, would st i l l be f u n c t i o n i n g and in the absence of any
extra di rect s t i mul ant s overall a c t i v i t y woul dn' t kindle i nt o t he
ki nd of neur al conf l agr at i on t hat can occur wi t h overdoses of
drugs l i ke cocaine and amphet ami ne.
Caf f ei ne' s rel at i v e saf et y is such t hat the only known deaths
at t r i but ed to c a f f ei ne overdoses have been acci dent s. The low-
est dose of c a f f e i ne known to hav e ki l l ed an adul t is 3,200
mi l l i gr ams, i nj ec t ed i nt o a pat i ent by a nur se who t hought the
syringe cont ai ned anot her drug (Gi l bert 1992). A f at al close of
caf f ei ne t aken by mont h would have to be at l east 5,000 mil-
l i grams: t he amount in about f or t y cups of st rong cof f ee con-
sumed very qui ckl y. Since t hi s quant i t y of cof f ee would i nduce
vomi t i ng long bef or e the l et hal dose was reached, i t ' s not likely
t hat even the most det er mi ned person could commi t suicide
wi t h cof f ee. In all cases of acci dent al deat h by caf f ei ne, the
act ual cause is not an ov cr s t i mi l l at i on of the brain but cardi ac
A Quicker Genius I 13 I
arrest induced by uncontrolled firing of the nerves that regulate
heartbeat.
Another consequence of caf f ei ne' s indirect action in the
brain is that i ngest i ng more of it doesn' t necessarily result in
greater st i mul at i on. If the levels of your exci t at ory neurot rans-
mi t t ers are low, no amount of caf f ei ne will boost them up, and
blocking more adenosine receptors with caf f ei ne will have l i t t l e
ef f ect . But there is anot her fact or at work here as well. Ani mal
experi ment s have demonstrated that caf f ei ne is biphasic, which
means that it has one ef f ect at low doses and other ef f ect s at
high doses. In one typical study using mice, caf f ei ne st i mul at ed
act i v i t y up to doses roughly equivalent to three to f our cups of
strong cof f ee. As the dosage increased above that level, activity
dropped. By the t i me dosage rose to the level equi v al ent to the
caf f ei ne content of ten cups of coffee, activity was lower than
it had been before the mice had ingested any caf f ei ne. In ot her
words, in mice at least, high doses of caf f ei ne act as a depres-
sant (Seale ei al . 1988). How mice actually feel af t er consuming
large amount s of caf f ei ne is, of course, unknown. But the re-
sults are i nt r i gui ng because they seem to indicate that caf f ei ne
is af f ect i ng more than j us t adenosine receptors in the brai n.
The current t heory is that high doses of caf f ei ne have a de-
pressant ef f ect because t he caf f ei ne i nt er f er es with another
molecular regul at or in the brai nan enzyme called phospho-
di est erase. By blocking the activity of this enz yme, very high
levels of caf f ei ne may set off a chemical chain of events t hat
inhibits neuronal fi ri ng.
The point of all this is t hat caf f ei ne doesn' t work the way
most people think it: works. The correlation between increasing
dose and increasing st i mul at i on applies only at doses equi v a-
lent to the caf f ei ne content of between one and f our cups of
coffee. Beyond t hat , pouri ng more caf f ei ne into the brain prob-
1321 Buzz
ably won' t i ncr ease st i mul at i onand it may hav e the reverse
ef f ect because of caf f ei ne' s act i ons on ot her mol ecul ar subsys-
t ems.
These new findings about caf f ei ne explain some of the or-
di nar y exper i ences of dr i nki ng coffee, tea, and ot her caf f ei nat ed
beverages. But t he whol e story is still not in hand. For one
t i l i ng, adcnosi ne i sn' t t he onl y neur ot r ans mi t t cr involved i n
sleep and waki ng. Sleep seems to restore many brai n circuits
and syst ems, not j u s t those usi ng adcnosi ne. Delaying or re-
duci ng sleep by us i ng c a f f e i ne may, t her ef or e, have subtle ef-
fect s on mood or cogni t i on t hat nobody has yet explored.
The Thi nk Dri nk?
The common experi ence of hav i ng one' s brai n "t urned on" by
c a f f e i n e l i as been v i v i dl y described by Nobel Pri z e-wi nni ng sci-
ent i st I ,con Cooper, a pr of essor at Brown U ni v er s i t y who is
deepl y i nv ol v ed in probi ng the br ai n. "This happens to me
every mor ni ng of my l i f e, " he t ol d wri t er George Johnson. "I
j u s t sit t her e at home wi t h my New York Times and my big pot
of tea, and a f t e r I h a v e enough c a f f ei ne in me 1 can j us t feel
my brai n going f r om a barel y consci ous level to t hi s high pitch,
as t hough I' ve t aken a dr ug. I' m suddenl y enormousl y awake
and v ery mani c, as you can sec. Ideas t umbl e out al most all
to be di scar ded by noon, unf or t una t e l y. But if I can focus on
somet hi ng wher e I r eal l y know the f act s, where I real l y under-
s t and t he probl em, t h a t ' s when somet hi ng mi ght happen"
( J ohns on 1991) .
Cer t ai nl y, Cooper i sn' t al one i n hi s experi ence. Through t he
cent ur i es, mi l l i o n s of people hav e used c a f f ei ne t o hel p spark
cr eat i v e t h o u g h t . One of t he more f a mous i s Johann Sebastian
Bach. Bach pas s i onat el y loved cof f ee, and in some ways no
mus i c bet t er capt ur es t he essence of the caf f ei nat ed exper i ence
A Quicker Genius I \ 33
than some of his more f r enet i c f ugues. Bach went so far as to
wr i t e a musical ode to his fav ori t e dr i nk, the Coffee Cantata,
a humorous one-act operetta about a st ern f at her ' s at t empt to
control his daught er ' s fondness for the bean. "Dear f at her , "
the girl implores at one point, "Don't be so strict! If I can' t
have my l i t t l e clemitasse of cof f ee three t i mes a day, I' m j ust
like a driecl-up piece of roast goat!"
Honorc dc Balzac, the great French writer, also loved coffee
and used it heav i l y. Me t ypi cal l y went to bed at 6:00 P. M. , arose
at mi dni ght , and wrote for twelve-hour stretches, dri nki ng cof-
fee cont i nual l y. "Coffee fal l s into your stomach and st rai ght -
away t her e is a general commotion," he wrote. "Ideas begin to
move like the bat t al i ons of the Grand Army on the battlefield
. . . things remembered arrive at f ul l gallop."
Philosophers, too, have t ur ned to caf f ei ne to f uel t hei r ru-
mi nat i ons. Immanuel Kant , Jean-Jacques Rousseau, and Vol-
t ai re all adored cof f ee. The Scottish philosopher James
MacKintosh even qui pped that "the powers of a man' s mind
are directly proportional to the quant i t y of coffee he drinks."
As i nt erest i ng as such examples are, of course, they in no
way prove t hat caf f ei ne act ual l y improves mental f unct i oni ng.
It coulci be that Bach, Balzac, and Voltaire would have been
j us t as creative wi t hout caf f ei ne. Even af t er decades of inves-
tigation i nt o caf f ei ne' s power to improve mental and intellec-
tual f unct i on, it is still far from clear t hat coffee really is the
"think dri nk. " No one doubts t hat caf f ei ne raises the overall
arousal level of the brain, delays the onset of sleep, and height-
ens al ert ness. But whether these ef f ect s translate into clearer
t hought , bet t er wri t i ng, or more cr eat i v i t y is an open quest i on.
At t he mol ecul ar level, the evidence is modestly encouraging.
For instance, c a f f ei ne arid other methylxanthines have been
shown to enhance long-term pot ent i at i ont he enduri ng syn-
aptic changes t hat arc thought to underlie memor y f or mat i on
134 I Buzz
(Tanaka 1990). Neurons bathed in modest l evel s of caf f ei ne
respond more vigorously to st i mul at i on and form longer-lasting
changes in t hei r connections with other neurons.
Another l i ne of research is also encouraging. Work from sev-
eral l abor at or i es has suggested t hat emotional arousal plays a
critical role in memory. Basically, our strongest memories are
of t hi ngs that arc emotionally provoking, in ei t her pl easant or
unpl easant ways. When adrenal i ne, the classic "fight-or-flight"
hormone, is released dur i ng such arousal, it seems to "prime"
the br ai n to remember t hi ngs in unus ual cl ari t y. It is theoriz ed
t hat this mechani sm evolved as nat ure' s way of ensuri ng that
animals clearly remember dangerous or provocative si t uat i ons
they encount er. This adr enal i ne connection to memory may
expl ai n the of t en- not ed phenomenon of people rememberi ng
exact l y what t hey were doing when they heard shocking news,
such as the fact t hat John F. Kennedy had been shot or that
the space shut t l e Challenger had exploded.
It is at least t heoret i cal l y possible t hat by st i mul at i ng emo-
tional arousal and, speci fi cal l y, by increasing levels of adrena-
line, c a f f e i n e may pri me t he brain t he same way t hat
provocative experi ences do. In one st udy, a 250-milligram close
of caf f ei ne raised adr enal i ne levels 207 percent and noradren-
aline levels 75 percent (Garcia 1993). The precise mechani sm
behi nd t hi s appar ent adr enal i ne boost remains unexplained.
Somewhat surpri si ngl y, the ef f ec t of caf f ei ne on memory for-
mation has not been well st udi ed in humans, t hough several
exper i ment s with ani mal s have yielded positive resul t s. In one
such st udy, rats given caf f ei ne bef or e l ear ni ng t hei r way
through a maze showed improved performance in l at er t ri al s
with the maz e t han did r at s given a placebo ( Bat t i g and Wetzl
1993).
As suggestive as such resul t s are, t hei r meani ng becomes elu-
sive when considered in light of the many st udi es of per f or -
A Quicker Genius I 135
mance conducted wi t h humans. Here the data are of t en
contradictory and di ffi cul t to i nt erpret . In general, the st udi es
have shown that caf f ei ne improves mental ability on tasks re-
quiring "speed," but degrades or has no effect on it with tasks
requi ri ng "power." For instance, caf f ei ne is hel pf ul in relatively
passive, aut omat i c, "data-driven" tasks such as audi t ory reac-
tion time, visual-choice reaction time, and per f or mi ng simple
arithmetic. It also improves the ability to attend to something
in a focused, sust ai ned way. But in st udi es of more complicated
tasks such as logical reasoning, numerical reasoning, readi ng
comprehension, and complicated arithmeticall of which re-
quire greater "central processing power"caffeine either has
had no detectable ef f ect or has actually degraded performance.
Such findings suggest that f amous coffee dr i nker s such as
Bach and Kant may have derived l i t t l e help from t hei r caf f ei ne
habits. And yet, despite scores of studies on the subj ect , it is
still too early to make such a j udgment . For one thing, the
t asks st udi ed to date hardl y capt ur e the range of activities en-
gaged in by humans; there is a great deal of di f f er ence between
numeri cal reasoning and philosophical synthesis. For another,
very few studies have taken into account the huge v ari at i on in
caf f ei ne response exhibited by i ndi v i dual human beings. In one
study in which these di f f er ences were considered, the research-
ers f ound indications that caffei ne' s ef f ect on mental perfor-
mance vari es with the impulsiveness of the user (Gilbert 1992).
Impulsive people, in this study, were defined as those who
tended to sacri fi ce accuracy for speed in v ari ous tasks and who
tended to be more aroused in the evening than in the morni ng.
When such people were given caf f ei ne in the morning (when
they were normally subdued), t hei r performance at tasks such
as proofreading for grammatical and typographical errors im-
proved. But when caf f ei ne was taken in the evening, their per-
formance was worse than when they had no caf f ei ne at all.
136 I Buzz
Conversely, people who scored low on i r npul s i v i t y t est s reacted
in exactly the opposite manner, scoring worse in the morni ng
and better in the evening.
In short, t here is no si mpl e answer to whet her caf f ei ne is, or
is not, hel pf ul in performi ng i nt el l ect ual t asks. The answer ap-
pears to depend both on the nat ur e of the task being consid-
ered and on the nat ur e of the person doing the task. The
gener al r ul e seems to be t hat c a f f ei ne is hel pf ul for people who
are not already nat ur al l y aroused and who are worki ng on rel-
atively st rai ght forward t asks t hat don' t r equi r e a lot of subtle
or abstract t hi nki ng. But even t hi s seemingly logical conclusion
must be t aken with a grain of sal t . Is wr i t i ng a st r ai ght f or war d
task? Is musical composition? Is comput er programmi ng? Per-
haps not for many people, and for folks who find such act i v i t i es
di f f i cul t caf f ei ne may, indeed, be more of a hi ndr ance t han a
help. But maybe for other people such "complex" t asks are
more aki n to play. Perhaps our notions of which tasks are
st rai ght forward and which arc complicated are too limited. It
may well be that the arousal i nduced by caf f ei ne produces re-
sults that are as uni que and di f f i cul t to predi ct as the actions
of any given i ndi v i dual .
If this is the case, then even the most rigorous research in
this area will be a poor gui de for any part i cul ar caf f ei ne user.
People t ryi ng to det ermi ne whether caf f ei ne helps or hur t s per-
f or mance in a given sphere of l i fe will be l ef t with no alter-
native but to become their own scientists and do their own
research. They will simply have to experi ment with t hei r fa-
vorite source of caf f ei ne, t ryi ng di f f er ent doses at di f f er ent
times in an ef f or t to sec what works for t hei r own uni que bio-
chemi st ry.
A Physical Drink
Early coffee and tea merchant s touted their wares as much for
the beverage' s t herapeut i c effect s on the body as for t hei r stim-
ul at i ng ef f ect s on the brai n. One 1657 adv er t i sement in a Lon-
don newspaper boasted t hat coffee was "a very wholesome and
physical drink t hat helpeth digestion, qui ckenet h the spirits,
maketh the heart lightsom, is good against eye-sores, coughs,
colds, rhumes, consumption, headache, dropsie, gout, and
scurvy. " Such promotional ent husi asm was seldom encum-
bered by logic: cof f ee was said to both whet and curl ) the ap-
138 I Buzz
pet i t e; to i ncrease al er t ness, but also to i nduce sleep; to cool
"hot" t emper ament s whi l e si mul t aneousl y wanni ng "cool"
t emper a ment s .
No one t oday views cof f ee or any ot her caf f ei ne- cont ai ni ng
dri nk as t hi s kind of medi ci nal panacea. But without a doubt,
c a f f ei ne a f f e c t s t he rest of t he body as power f ul l y as i t af f ect s
the br ai n. It al t er s the f unct i oni ng of nearly every bodily sys-
tem, f r om the blood vessels in the head to the muscles con-
trolling the toes and ev er yt hi ng in between. For most people,
t hese side ef f ect s are mi nor annoyances and are irrelevant to
t he desi red ef f ect of ment al s t i mul at i on. But caf f ei ne' s side
ef f ect s arc act ual l y the mai n poi nt for ot her s. Many at hl et es,
for i nst ance, t ake caf f ei ne t o boost t hei r physical per f or mance,
and t her e are peopl e who consume c a f f ei ne t o curb t hei r ap-
petite, to lose: wei ght , or as an i nexpensi v e l axat i v e.
Ca f f e i ne owes i t s r emar kabl y broad reach t o t wo related
f act s . First of a l l , t he br ai n cont rol s many bodily f unct i ons,
ei t her di r ect l y v i a nerve i mpul ses or i ndi r ect l y v i a hormones.
By al t eri ng t he br ai n, c a f f e i n e aut omat i cal l y al t ers al l systems
r egul at ed by the br ai n. Ca f f e i n e also di rect l y af f ect s many part s
of the body by at t achi ng to adcnosi nc receptors f ound out si de
the br ai n. Here, as i n t he br ai n, c a f f ei ne sometimes causes an
accel erat i ng or t ensi ng response. It causes the heart to beat
more r api dl y, it constricts some blood vessels, and it causes
certain t ypes of muscles to contract more easily. But, paradox-
ically, c a f f ei ne can also cause a rel axat i on response. It can, for
i nst ance, r el ax the ai rways of the lungs and cause certain types
of blood vessels to open.
These cont radi ct ory ef f ect s can be explained by probing
deeper into the subj ect of adenosine receptors. To t hi s poi nt ,
we've dealt with only one type of adenosi ne recept ort he kind
in the br ai n t hat cause neur ons to slow down. But t here are at
least three t ypes of adenosi ne receptors, all of which di f f er in
The Body, Wired I 139
the signals t hey send a f t e r adenosine binds to them. A
t
recep-
tors, the kind we' ve been dealing with thus f ar , quiet act i v i t y
by i nhi bi t i ng cell firing. But A
2
receptors set off a di f f er ent
biochemical cascade when adenosine binds. This cascade sends
j us t the opposite message: it excites neurons to fire. The chem-
ical signals sent by A, receptors, the most recently i dent i fi ed,
are not yet underst ood (Salvatore 1993).
For our purposes, it is not part i cul arl y i mport ant to know
whi ch ki nd of receptor is responsible for which bodily reaction,
though the issue is of great i nt erest to drug makers who hope
to find molecules that bind selectively to j us t one type of re-
ceptor. Here it' s enough to underst and t hat adenosine recep-
t ors come in several v ari et i es and that t hi s is the f undament al
reason t hat adenosi neand t hus caf f ei nehav e such appar-
ently contradictory ef f ect s in di f f er ent parts of the body. This
chapt er will exami ne some of caffei ne' s most i nt erest i ng re-
percussions, from the promotion of uri nat i on to the elimina-
tion of headaches.
The Enfeebling Liquor
Unlike alcohol, caf f ei ne is seldom viewed as an aphrodi si ac.
More t ypi cal l y, coffee, tea, and other caf f ei nat cd dri nks are
associated with workpart i cul arl y work involving thinking,
reading, wr i t i ng, or t al ki ng. Although such intellectual act i v i t i es
might for some people const i t ut e sexual forepl ay, this isn' t
their usual role and caf f ei ne is not generally associated with
sex in popul ar cul t ur e. Quite the opposite. Since its introduc-
tion in Europe, caf f ei ne has more of t en been linked to celibacy
at best, impotence at worst. The two most popul ar legends
about the origins of coffee and tea involve using these sub-
stances to achieve a greater communion with Godhardly a
ribald a c t i v i t y. The followers of Bodhi dharma, remember, used
140 I Buzz
tea to enhance medi t at i on, and the monks of Meeea used Kal-
di ' s di scov ery of coffee t o ext end t h e i r abi l i t y t o pr ay.
In 1764, a broadside appeared on the streets of London:
"The Women' s Petition agai nst Cof f ee, Represent i ng to Pub-
lick Consi der at i on t he Grand Inconveniences accrui ng t o ( hei r
SEX from t he Excessive use of t hat Dryi ng, Enfeebl i ng LIQ-
UOR" (Schi v cl busch 1992). The aut hor s of the tracta group
cal l i ng i t sel f the Keepers of the Li bert y of Venuswere af r ai d
t hat coffee would make "men as u n f r u i t f u l as those deserts
whence t hat unha ppy berry is said to be br ought . "
Despite t hi s apparent concern for the sexual health of men,
t he pet i t i oner s may hav e had an ul t er i or motive. Cof f eehouses
at t hat time were spri ngi ng up by the t housands, and t hey were
usual l y men- onl y est abl i shment s. The women suppor t i ng t he
pet i t i on may have been ai mi ng a clever at t ack on di scri mi na-
tory cof f eehouses by h i t t i n g at a classic mal e weak spot. None-
t hel ess, v ari at i ons on the "enfeebl i ng" t heme cont i nued to be
played. ' I ' h e ni net eent h- cent ur y poet - hi st or i an Jul es Mi chel ct
described cof f ee as "ant i cr ot i c. " "Coffee, " he said, "at l ast re-
places sexual ar ousal with s t i mul a t i on of the intellect." Mieh-
elet' s vi ew was har dl y i di os yncr at i c. At t hat t i me, cof f ee was
oft en recommended as the pr ef er r ed dr i nk of clerics and ot her
cel i bat es.
Ca f f e i ne received no better press in the t went i et h cent ur y.
In a 1931 book about narcot i cs and s t i mul a nt s , L. Lcwin wrote
t hat coffee coul d "st eri l i z e nat ur e and ext i ngui sh carnal de-
sires." Today nobody views cof f ee as "st eri l i z i ng. " But nei t her
does cof f ee or any ot her caf f ei ne- cont ai ni ng beverage connote
sexual v i r i l i t y or physi cal passi on. The romance and myst i que
s ur r oundi ng cof f ee and tea r emai n pr i mar i l y i nt el l ect ual . If pas-
sion is involved, it is a passion for i deas and for t hei r lively
exchange unde r t he i nf l uence of a drug t hat encourages qui ck
t h i n k i n g and wit.
The Body, Wired I 141
It would be sat i sf yi ng, at this point, if modern science could
be called on to of f er evidence that would either subst ant i at e
or rej ect these centuries-old views. Unf or t unat el y, such dat a do
not exist. Unlike alcohol's effect s on human sexual response,
which has been the focus of hundreds of papers over the years,
the l i t er at ur e on caf f ei ne and sex is anemic to say the l east . If
the subj ect is mentioned at all in scholarly j our nal s or books,
it is simply to note, as one aut hor did, t hat "no specific ef f ect
of caf f ei ne on sexual f unct i on has been demonst rat ed" (Hoi-
lister 1975).
The only aspect of sex that has been examined in any detail
with respect to caffei ne is rat her far removed from the act
i t sel f. Several studies have demonst rat ed that sperm exposed
to caf f ei ne swim f as t er and more energetically than normal. In
short, they wiggle more, allowing them to penetrate cervical
mucus more readily (James 1991). The upshot is an increased
chance of fert i l i z at i on if an egg is nearby. These findings led
some researchers to seriously consider the use of caf f ei ne for
human in v i t r of er t i l i z at i on t herapy, since one of the big prob-
lems with sperm stored at ext remel y low t emper at ur es is that
when thawed they have "lowered motility. " Unfort unat el y,
however, the caf f ei ne concentrations required to kick-start
sperrn in a glass dish are more than a t housand t i mes higher
than the level of caf f ei ne in the body following even large doses
of coffee. In fact, the 1,500 milligrams per l i t er concentrations
used in some experi ment s may induce chromosomal damage,
making this method of enhancing f er t i l i t y dubious indeed.
The ef f ect of caf f ei ne on sperm mot i l i t y is the only aspect
of human sexual i t y that appears to have been examined in any
rigorous way. One possible explanation for this is that caffei ne' s
ef f ect on sex is so t ri v i al that nobody has felt mot i vat ed to go
to the trouble and expense of measuri ng it. Indeed, caf f ei ne' s
rel at i onshi p to sexual response may be so slight that i t ' s
f 42 / Buzz
swamped by the much st r onger variable we met earl i er: the
mi nd. We saw t hat people' s expect at i ons about alcohol can
override the si gni fi cant physiological ef f ect s exerted by t hat
drug. It is likely t hat when it comes to caf f ei ne and sex, people
feel what t hey expect to feel : expect caf f ei ne to be an aphro-
disiac, like alcohol, and it probably will be; expect it to snuf f
out your carnal desires, and it will probably do j us t t hat .
Legal Speed
In December 1993, things looked good for t wcnt y-four-year-ol d
Sylvia Gerasch, the Eur opean champion in the 100-meter
breast stroke. But only a mont h l at er she was st ri pped of her
swimming title, dropped f r om the German raci ng team, and
banned f r om f ur t her compet i t i on for t wo years ( Reut er s 1994).
The r eason? She dr ank too much cof f ee.
In Januar y 1994, as part of the r out i ne drug t est i ng of ath-
letes, Gerasch was f ound to have 16 micrograms of caf f ei ne per
milliliter of blooda level consi derabl y higher t han the offi ci al
l i mi t of 12 micrograms per mi l l i l i t er. Offi ci al s of the European
Swimming Federat i on pointed out t hat such levels could be
achi eved onl y if Gerasch had consumed the equi val ent of about
eight cups of st rong cof f ee in a short period prior to the t est i ng.
Gerasch protested the rul i ng, saying t hat she was only dr i nki ng
her normal amount . But she di dn' t say what t hat amount was,
and t he of f i ci al s stood f i r m.
Gerasch is har dl y an isolated case. Other swimmers, r unner s,
and bicycle racers have been penalized in recent years for ex-
cessive use of caf f ei nea pract i ce t hat for many years has been
widespread but not always regul at ed. The issue first came to a
head in 1962, when the I nt er nat i onal Olympic Committee
listed caf f ei ne as a "doping agent "t hat is, a subst ance (such
as steroids) t hat provi des an art i fi ci al per f or mance boost. In a
The Body, Wired I 143
controversial move, the committee removed caffei ne f r om the
list in 1972, but then r ei nst at ed it in 1984. Although this action
surpri ses people who don' t consider caf f ei ne a drugmuch less
a "doping agent "current research strongly supports the
Olympic Committee' s act i ons.
Caf f ei ne has long been used to increase physical endurance,
in both humans and animals. In Tibet, horses and mules work-
ing at ext remel y high elevations are of t en given large vessels of
tea to increase t hei r capacity for work (Gilbert 1992). The an-
imals' mast ers, too, keep themselves going with caf f ei ne. The
di st ances between Tibetan villages is sometimes reckoned by
the number of cups of tea necessary to sustain a person trav-
eling that route, three cups of tea being roughly equal to 8
ki l omet ers.
Studies of caf f ei ne' s ef f ect s on athletic per f or mance have,
for the most par t , corroborated such anecdotal observations.
For i nst ance, recent research at Christ Church College in Can-
t erbury, England, showed that caf f ei ne reliably increased per-
f or mance in several runni ng events. Eighteen runners of
various abilities ran a 1,500-meter time t r i al on two occasions:
first af t er dr i nki ng two cups of coffee containing a total of 300
to 350 milligrams of caf f ei ne, and then af t er dri nki ng the same
amount of decaffei nat ed coffee. Fourteen of the eighteen sub-
jects ran f as t er a f t er ingesting the caf f ei ne. On average, their
times improved by f our seconds. Another test measured caf-
fei ne' s ef f ect on "burst" act i v i t y: the ability of ten competitive
r unner s to "kick" the final 400 met ers of a 1,500-meter time
trial. The r unner s covered the first 1,100 meters at a predet er-
mined pace and were then instructed to run the final leg as
f ast as they could. All ten "kicked" f ast er on caf f ei neand also
produced less f at i gue- i nduci ng lactic acid in t hei r blood.
In anot her study, bicyclists who took 330 milligrams of caf-
feine one hour before exercising were able to pedal an average
144 I Buzz
of 19.5 percent longer t han subject s who dr ank no caf f ei ne
( Bur ke 1992) Impressed by r esul t s such as these, many profes-
sional bi cycl i st s dr i nk caf f ei ne- cont ai ni ng beverages both be-
f or e and dur i ng a race. Some even insert caf f ei ne supposi t ori es
bef or e a race in an at t empt to provi de a sust ai ned dose with
no stomach upset .
Exact l y how caf f ei ne boosts physical per f or mance is still not
cl ear. Attention originally focused on the way caf f ei ne af f ect s
muscl e cont ract i on and r ef l exes. In 1913, the scientist Storm
van Leeuwen report ed t hat caf f ei ne i ncreased spinal refl exes i n
cats. Cont emporari es reported si mi l ar increases in the f ami l i ar
"kneecap, " or pat el l ar, refl ex in humans . But these and ot her
early studies could not be replicated later (Battig and Wetzl
1993). More recent st udi es of i ndi v i dual muscle fibers isolated
out si de the body show t hat caf f ei ne can increase the speed and
f or ce of cont r act i on, but only when caf f ei ne levels are signifi-
cant l y higher than those f ound in the blood of even the most
caf f ei ne- abusi ng at hl et e (Fredholm 1984). A rel at i vel y new line
of research suggests that muscle cont ract i ons may be st i mu-
lated by dopami ne released by c a f f ei ne in the br ai n (Jossclyn
and Beninger 1991). But t hi s t heory is still qui t e sketchy and
t ent at i v e. In t he meant i me, at t ent i on among sports physi ci ans
has t ur ned away f r om the muscles t hemsel ves and toward the
f uel t hat powers muscul ar act i v i t y i n general .
Many st udi es have f ound t hat caf f ei ne releases fat stored i n
cells and breaks it down i nt o the smal l er f at t y- aci d chains t hat
the body bur ns as f ue l . Caf f ei ne may release these f a t t y acids
directly via some as- yct - unknown process, or it may do so in-
di r ect l y by r ai si ng adr enal i ne levels. Regardless of the mecha-
ni sm, caf f ei ne' s abi l i t y to liberate some of the f uel suppl y
stored in fat may explain its benefi ci al ef f ec t s on at hl et i c per-
f or mance.
This ef f ect can be strongly i nf l uenced by diet. One st udy
'The Body, Wired I 145
f ound that t he gr eat est release of f a t t y acids occurred a f t er
at hl et es ate a r at her atypical meal of sausage, bacon, and eggs
(Weir et al. 1987). The ef f ect was negligible when athletes ate
a more common "carbohydrate loading" br eakf ast of cereal,
toast, and orange j ui c e. It has also been f ound t hat peak f r ee-
fat t y-aci d l ev el s occur t hree t o f our hours a f t er caf f ei ne inges-
t i on. This suggests t hat c a f f ei ne may be acting on more than
j us t fat l i ber at i on, since it has been shown to boost per f or -
mance in both shor t - dur at i on and endur ance sports. It may be
that athletes engaging in shor t - dur at i on events, for instance,
benef i t f r om caf f ei ne' s st i mul at i on of the cent ral nervous sys-
tem, whi l e endurance athletes may get t hei r boost f r om the
longer-term i ncrease in f at t y acids released by caf f ei ne af t er
several hours.
Whether c a f f e i ne hel ps or hur t s a given athlete is a mat t er
of i ndi vi dual response. Some at hl et es find it hel pf ul , while oth-
ers find t hat they s u f f e r f r om an acid stomach, increased nerv-
ousness, or dehydrat i on and t hus avoid it. In any case, the
cur r ent legal l i mi t s for c a f f ei ne are s uf f i ci ent l y high to allow a
wide l at i t ude of exper i ment at i on. The Olympic legal limit of
12 mi crograms per mi l l i l i t er (which has been widely adopted
for ot her sport i ng events) is equi v al ent to about six cups of
coffee consumed wi t hi n t hi r t y mi nut es . Since t he r esul t s ob-
tained in the st udi es ment i oned earlier were f ound at levels
consi derabl y below this l i mi t , it appears t hat caf f ei ne is likely
to cont i nue to be one of the most widely used doping agents
in athletic competitions.
Liquid Diet
The active ingredient in most over-the-counter diet aids used
to be not hi ng more than a 200-milligram close of caf f ei ne
the equi val ent of about two cups of cof f ee. Such pills were a
146 I Buzz
gold mi ne for dr ug compani es because c a f f ei ne is a r el at i v el y
i nexpensi ve raw mat er i al and customers are willing to pay
handsomel y for anyt hi ng promi si ng easy weight loss. But i n
1991, the Food and Drug Admi ni st r at i on banned caf f ei ne from
al l wei ght -reduct i on products, rul i ng t hat caf f ei ne nei t her sup-
presses appet i t e nor di rect l y causes weight loss. (The compa-
nies sel l i ng t he pills pr ompt l y replaced c a f f ei ne with anot her
st i mul ant : phenyl propanol ami ne, one of the mildest members
of the large amphet ami ne f ami l y. ) Despite the FDA r ul i ng
about caf f ei ne, however, many people still believe that a cup
of joe will help them lose wei ght . Int erest i ngl y enough, caf f ei ne
does have some demonst rabl e ef f ect s t hat , at least t heoret i cal l y,
could help dieters. The problemas the FDA recognizedis
t hat for most people t hese ef f ec t s are mi ni mal to the point of
i nsi gni fi cance.
We j us t saw one ef f ect t hat might seem t o j u s t i f y t he di et er' s
belief in caf f ei ne: its abi l i t y to release fat and break it down
into us ef ul f a t t y aci ds. This i s pot ent i al l y si gni f i cant for at hl et es
because t hey are exer ci si ng so har d t hat t hei r muscles readi l y
bur n t he l i berat ed f a t t y aci ds. But for more sedent ary types,
the f a t t y acids released by a cup or two of coffee are l i kel y to
simply be reconverted to fat once caf f ei ne levels drop. Caf f ei ne,
in other words, i sn' t a "fat burner, " but a "fat releaser." Ex-
ercise is still needed to e f f e c t wei ght r educt i on. Caf f ei ne by
i t sel f simply sets the st ageand even t hen the ef f ect is rel a-
tively smal l and of use pr i mar i l y to at hl et es for whom even a
2 percent i ncr ease in energy av ai l abi l i t y might prove to be the
wi nni ng margi n.
What about anot her f act or somet i mes t out ed as being sig-
ni f i cant to di et er s: caf f ei ne' s abi l i t y to rev up the body' s meta-
bolism? Several st udi es have f ound t hat moderat e consumpt i on
of caf f ei ne does rai se the basal met abol i c r at epr obabl y by
slightly i ncreasi ng a dr ena l i ne l ev el sand that this resul t s in a
The Body, Wired I 147
small elevation of body t emper at ur e and corresponding in-
crease in calorie consumpt i on. This rise, however, is very small
indeed. Over the course of a day, the average i ncrease in calorie
consumpt i on is between fifty and a hundr ed calories ( Bat t i g
and Wetzl 1993). Given a per f ect l y uni f or m diet, t hi s could
have an ef f ect since even a small increase adds up over the
long haul . But in real i t y, people' s diets are far from uni f or m
and readi l y respond to variablesnot the least of which is hun-
ger i t sel f t hat are far more powerful t han the modest tem-
perat ure-rai si ng ef f ect s of caf f ei ne.
What about the possibility t hat caf f ei ne curbs hunger by
i nt er f er i ng wi t h the brai n' s appet i t e-cont rol center? Again,
there may be something to this idea, but the necessary studies
on humans hav en' t been done. It is theoretically possible that
caf f ei ne somehow inhibits parts of the brai n, such as the hy-
pot hal amus, involved in regulating appet i t e. But nobody has
yet suggested a plausible mechanism for such inhibition, and
ani mal st udi es have f ound no rel at i on between caf f ei ne and
food i nt ake. It ' s unl i kel y, t her ef or e, t hat any of these caf f ei ne
side ef f ect s plays a si gni fi cant role in weight loss. In f act , for
some people, caf f ei ne may act ual l y make it harder to eat a
balanced, healthy diet. A st udy by researchers at the Uni v ersi t y
of Michigan f ound t hat of 171 pat i ent s wi t h eat i ng di sorders,
those who consumed a lot of caf f ei ne (more t ha n 750 milli-
grams, or the equi v al ent of about eight cups of cof f ee a day)
binged, f a s t e d, and used l axat i v es and diet pills more of t en,
and were more likely to smoke and abuse alcohol, t han pat i ent s
whose i nt ake was moderat e (Li v ermore 1991).
This is not the same as saying caf f ei ne causes increased bi nge
eat i ngj ust t hat t here is an association between t he two. It
could well be that what ever underl i es the pat t er n of ext r eme
di et i ng also suppor t s high use of caf f ei ne. But the finding has
renewed i nt erest i n t he r el at i onshi p between c a f f ei ne and i n-
148 / Buzz
gcstivc behaviors and is of i nt erest to those st udyi ng eat i ng
di sorders.
Tremors and Twitches
Ca f f ei ne' s abi l i t y t o i ncrease r epet i t i v e or i nv ol unt ary muscle
mov ement s has received l i t t l e sci ent i fi c at t ent i on. A common
exampl e of such an e f f e c t i s "t r eadl e f oot "t he rapi d jiggling
or pumpi ng of one or both legs, somet i mes observed in people
who have consumed c a f f ei ne and are si t t i ng down. Another
wel l -known ef f ect of caf f ei ne is increased hand t remor, which
has been measured in numer ous experi ment s. And some peo-
ple find that c a f f ei ne increases the number and i nt ensi t y of
small muscl e t wi t ches in such pl aces as the eyelids, arms, or
legs.
The mechani sm behi nd these phenomena is st i l l not well
under st ood. The muscles involved in such t remors and
t wi t ches arc skel et al , as opposed to cardiac or smooth muscle.
Some exper i ment s on skeletal-muscle st ri ps have demonst rat ed
t hat c a f f ei ne increases cont r act i ons, which might seem to ex-
pl ai n some of the t wi t ch phenomena. But the concent rat i ons
of caf f ei ne r equi r ed to produce such cont ract i ons are almost a
hundr ed t i mes hi gher t han levels f ound i n people's blood af t er
ingesrion of moderat e a mo u n t s of caf f ei ne.
Anot her pos s i bi l i t y i s t hat caf f ei ne af f ect s skeletal muscles
i ndi r ect l y. Exper i ment s on frogs show t hat caf f ei ne can release
acet yi chol i ne, t he neur ot r ans mi t t er responsible f or i ni t i at i ng
muscl e cont r act i on.
Regardl ess of the causat i v e agent, the t r emor s and twitches
experi enced by some users of c a f f ei ne are usual l y harmless.
St i l l , some heavy dr i nker s of caf f ei nat ed products experience
cardi ac a r r h y t h mi a s ( i r r egul ar heart beat s) or pal pi t at i ons (a
The Body, Wired I 149
fluttering hear t beat ), both of which, while not usual l y lethal,
could be problematic for people wi t h chronic heart conditions.
Prompting Nature's Call
One of the most well-known (if least-discussed) side ef f ect s of
c a f f ei ne is its st i mul at i on of ur i nat i on and defecat i on. Like al-
cohol, caf f ei ne increases ur i nat i on both directly and i ndi rect l y.
Since caf f ei ne is usually consumed in beverages, the l i qui d by
i t sel f will result in an increased urge to uri nat e. But the kidneys
are also rich in adenosine receptors. Adenosine helps regulate
the delicate balance between blood flow and ur i ne out put .
When caffei ne blocks these receptors, blood vessels dilate, in-
creasing the filtration rat e and producing more uri ne. Both caf-
fei ne and alcohol t hus are mild di uret i csdrugs t hat increase
uri nat i ont hough the two substances achieve this end by di f -
f er ent means. Caf f ei ne' s di ur et i c ef f ect i s usual l y mild and
harml ess. But for athletes and others who are likely to perspire
heavily, excessive consumpt i on of caf f ei ne could lead to de-
hydrat i on.
Caf f ei ne is also a l axat i v e. Like the kidneys, the colon is well
endowed with adenosine recept ors. Here adenosine helps con-
trol the tone of the smooth muscles used to propel feces on
its way. Mere, too, adenosine is the signaling molecule used to
mai nt ai n the balance between r el axat i on and cont ract i on. But
unlike the act i on in the ki dney, where caf f ei ne causes a dila-
t i on, in the colon it causes a constriction. When adenosine
receptors in the colon are blocked by caf f ei ne or other meth-
yl xant hi nes, the nor mal relaxation messages are blocked. The
smooth muscles t hus more easily contract in a char act er i st i c
rhyt hm called i nt est i nal per i st al si s. Even moderat e closes of caf-
f ei ne can set off t hi s peri st al si s whet her or not the body was
I SO I Buzz
ready to dispose of its feces. Although in some cases caf f ei ne
can cause di ar r heat ypi cal l y among those who seldom con-
s ume c a f f ei na t c d bev eragesno harm from t hi s l axat i v e ef f ect
has ever been r epor t ed.
Of Headaches and Pai nki llers
Headaches a f f l i c t millions of people every day. The vast ma-
j or i t yabout 90 per cent ar e caused by excessive tension in
the head and neck muscl es. But about 8 percent are v ascul ar
headaches, caused by the excessive dilation of blood vessels in
the br ai n. Migraines are a par t i cul ar l y i nt ense ki nd of v ascul ar
headache. Hangover headaches are another t ype.
The di amet er of cerebral blood vessels is regulated by
smooth muscles, which, in t urn, are controlled by adenosine.
When adenosine levels rise, blood vessels relax and open up.
Blocking adenosi ne receptors with caf f ei ne negates this ef f ect ,
causi ng vessels to const r i ct , ' [' hi s vasoconst ri ct i on is relatively
mi nor and us ual l y goes unnot i ced, but for people suf f er i ng
f r om v ascul ar headaches the reduct i on of cerebral blood flow
is welcomed. This is why caf f ei ne is the active ingredient in
many prescri pt i on and nonpreseri pt i on migraine t reat ment s.
One of the bet t er-known brands, Cafergot, contains 100 mil-
l i grams of c a f f e i ne in each t abl et or supposi t ory.
But as a headache remedy, caf f ei ne can be tricky to manage.
As we wi l l see in more detail in the next chapter, the body
adapt s qui ckl y to caf f ei ne. One adj us t ment is that the muscles
controlling t he cerebral blood vessels increase t hei r rel axat i on
response to compensat e for the increased constriction caused
by c a f f e i ne . This adapt at i on goes unnot i ced as long as caf f ei ne
levels are const ant l y repl eni shed. But if c a f f ei ne i nt ake is sud-
denly st opped, the i ncreased rel axat i on is no longer count er-
balanced by c a f f e i ne . The vessels dilate much more t han
The Body, Wired I 151
normal, and a t hrobbi ng "rebound" headache can ensue. Head-
aches, in f act , are the most common symptom of caf f ei ne with-
drawal. This potential for rebound headaches and the general
di f f i cul t y of accurat el y sel f - admi ni st er i ng ef f ect i v e doses of caf-
fei ne are t he pr i mar y reasons t hat migraine s uf f er er s are ad-
vised to abstain f r om caf f ei ne, even though it may bring
t emporary relief dur i ng an at t ack.
People s uf f er i ng f r om headaches brought on by wi t hdrawal
from caf f ei ne of t en reach i nt o t hei r medicine cabinet for some
aspi ri n or a non-aspi ri n pain reliever like Tylenol or Mot ri n.
Taking the medi cat i on, they may find that t hei r headache van-
ishes and they at t r i but e t hei r relief to the aspi r i n, acetamino-
phen, or i buprofen in the tablets. But in f act , they may simply
have given themselves a classic "hair~of-the-dog" cure. Many
brands of aspi ri n and aspi ri n subst i t ut es include a si gni fi cant
dose of caf f ei ne. The usual level is 65 milligrams per tablet,
which means that the st andard two-tablet dose admi ni st ers a
caf f ei ne jolt equal to t hat in a cup of v ery strong cof f ee or two
cups of black tea. Thi s addi t i on of caf f ei ne could be cynically
i nt er pr et ed. One might suspect that drug companies add caf-
f ei ne to t hei r pai n-rel i ev er product s because they recognize
t hat headaches are of t en caused by caf f ei ne wi t hdrawal and
they under s t and t hat the easiest way to relieve drug withdrawal
symptoms is by the admi ni st r at i on of the drug i t sel f .
But t her e' s a more honest and i nt er est i ngexpl anat i on for
the presence of caf f ei ne in pai nki l l ers. For reasons t hat r emai n
obscure, caf f ei ne si gni fi cant l y increases the analgesic ef f ect i v e-
ness of both aspi r i n and aspi ri n subst i t ut es such as acetamin-
ophen. Data from more t han t hi r t y clinical t ri al s involving
more t han 10,000 pai n pat i ent s unequi v ocal l y support t hi s eon-
elusion. Most of these pat i ent s had pain f r om post part um ut er-
ine crampi ng or cpisiotomies, but some had pain f r om oral
surgery, headache, or cancer. On average, when the pai nki l l ers
152 I Buzz
were given without c a f f e i n e , the doses had to be about 40 per-
cent larger to obt ai n the same degree of rel i ef as was obtained
when the analgesics were given with c a f f e i n e (Laska et al.
1984). It is considered s af er to give caf f ei ne suppl ement s to
pat i ent s who arc get t i ng l arge dai l y closes of pai n r el i ev er s t han
to give t hem the even higher closes of medi ci ne that would be
r equi r ed wi t hout t he c a f f ei ne.
How c a f f ei ne exert s t hi s hel pi ng hand is a myst ery. One the-
ory i s that t he well-being, al ert ness, and decreased i r r i t abi l i t y
i nduced by caf f ei ne somehow count er act s the perception of
pa i n. Some evidence also suggests t hat caffei ne' s i nt er f er ence
with adenosi ne receptors t hr oughout the body i nhi bi t s the pro-
duct i on or release of chemicals t hat cause pai n and i nf l am-
mat i on. At t he moment , nobody is wor ki ng to pin down the
exact mechani sm because so many st udi es have v er i f i ed the
ut i l i t y of addi ng c a f f ei ne to pai n ki l l ers t hat it is seen as a moot
poi nt .
Caffeine and PMS
In the mid-1980s, when she was an a s s i s t a nt professor at Tuf t s
Uni v er si t y, Annet t e MacKay Rossignol was involved i n t abu-
l at i ng the resul t s f r om a general sur v ey of st udent heal t h and
di et ar y ha bi t s . While pori ng t hr ough t he dat a on heal t h- r i s k
behav i or s f r om exerci se to seat-belt use, Rossignol detected an
appar ent pat t er n between c a f f ei ne and pr emenst r ual syndrome,
j u s t for f u n , she ran t he number s. The r es ul t was a correl at i on
st rong enough to pi que Rossignol' s i nt er est and l aunch a re-
search agenda that l ast ed for many year s.
Pr emenst r ual syndrome is a const el l at i on of sympt oms t hat
can i ncl ude i r r i t a bi l i t y, anxi et y, depressi on, breast swelling and
t ender ness, fl ui d r et ent i on, food cravings, and headaches. Ros-
signol not i ced t hat women wi t h moderat e to severe PMS
The Body, Wired I 153
tended to consume the most caf f ei ne, of t en in the f or m of sof t
dri nks. Over the next eight year s, she and vari ous colleagues
conducted a series of ever more refi ned studies in an at t empt
to pin down the r el at i onshi p between c a f f ei ne and PMS. She
even conducted a st udy in China to i nv est i gat e the i nci dence
of PMS among women working in a tea f act or y who consumed
relatively large amount s of caf f ei ne. In t hi s and all of her other
studies, she f ound t hat the women s uf f er i ng the most severe
PMS also tended to consume the most caf f ei ne- cont ai ni ng bev-
erages, whet her tea, soft dr i nks, or cof f ee. Rossignol recognized
that there might be al t er nat i v e expl anat i ons for the associa-
t i onfor instance, t hat the ef f ect she was seeing was due not
to caf f ei ne but to the ext r a fluid consumed by t he women
drinking a lot of caf f ei ne- cont ai ni ng beverages. But a 1990
study cont rol l i ng for such an ef f ect still found a cl ear r el at i on
between caf f ei ne and PMS.
Work by John W. Phillis of Wayne State Uni v er si t y has sug-
gested a possible expl anat i on for Rossignol' s findings. Phillis
(1989) has f ound t hat two of the reproduct i ve hormones that
fluctuate mont hl y in women di rect l y af f ect adenosine levels in
the brai n. The two hormones exert opposite ef f ect s . Beta-
estradiol appears to mimic caf f ei ne' s ef f ect s: it ant agoni z es ad-
enosine' s na t ur a l i nhi bi t or y e f f e c t s and t hus produces a ki nd
of mild st i mul at i on t hat some women charact eri st i cal l y at t r i b-
ute to the l at e fol l i cul ar phase of the menst rual cyclethat is,
the phase j ust before ov ul at i on. Conversely, the hormone pro-
gesterone enhance* adcnosine' s actions. This may expl ai n some
of the fat i gue and mood depression commonly reported by
women i n the week a f t e r ov ul at i on as progesterone l ev el s i n-
crease dr amat i cal l y. Pl ummet i ng progest erone levels j us t bef or e
menst ruat i on may, somewhat paradoxically, account for the
feelings of tension, i rri t abi l i t y, and anxi et y t hat are sometimes
reported duri ng t hi s t i me. Since caf f ei ne exert s i t s ef f ect s by
I S4 / Buzz
bl ocki ng adcnosi ne, i t coul dat l east t heor et i cal l yi nf l uence
the way these hormone fluctuations af f ect a woman' s physical
and ment al well-being.
If t he t heori es j us t out l i ned are correct and i t i s not yet
clear t hat t hey are correct t hen women who want to use caf-
f ei ne to hel p cont rol the emot i onal f l uct uat i ons of PMS should
modul at e t hei r c a f f ei ne consumpt i on t o count er act progester-
one' s mild depr essant ef f ec t s . About three to f our days before
t hey menst r uat e, women shoul d stop caf f ei ne use altogether
and abstain f r om eaf f ei nat ed beverages right through t hei r
mens t r uat i on. Af t er me n s t r u a t i o n has fi ni shed, they could ei -
ther r emai n eaf f emc- f r cc or dri nk onl y small amount s unt i l
they ovul at e again. Rossignol and Phillis f ound t hat many
women did change t hei r caf f ei ne consumpt i on over the course
of a menst r ual cycle, but not in the way they expected (Ros-
signol et al . 1991). Women drank the most caf f ei ne dur i ng
their mens t r ual fl ow, drank less as ov ul at i on approached, and
then increased t hei r c a f f ei ne consumpt i ona pat t er n almost
exact l y t he opposite of t hat suggested to be benef i ci al .
Rossignol and Phillis t hcor i / c t hat t hi s pat t er n of caf f ei ne
use may reflect an unsuccessf ul at t empt by the women to sel f-
medi cat e with c a f f ei ne. Women may respond to the unpleas-
ant feel i ngs associated with the peak progest erone levels by
i ncr easi ng t hei r c a f f e i n e i nt ake. They mai nt ai n t hi s level right
t hr ough t he me ns t r ua l fl owat exactly t he time, t he Rossignol
and Phillis st udi es i ndi cat e, t hey should abst ai n f r om caf f ei ne.
This may act ual l y make t hei r sympt oms worse by l eadi ng t hen,
t o dr i nk even more c a f f ei ne i n an at t empt t o overcome t hei r
unpl eas ant sympt oms.
As i nt er es t i ng and suggestive as t hi s work is, it should be
weighed agai nst one' s per sonal experi ence. The phenomenon
of PMS i t sel f is st i l l not clearly underst ood, let alone the re-
l at i onshi p bet ween c a f f e i ne and r epr oduct i v e hormones. Al -
The Body, Wired I 155
though the association Rossignol f ound is well established, it
does not prove causat i on: some ot her f act or may be at work
that is unr el at ed to c a f f ei ne. The crucial study, in fact , has yet
to be done. This would be a controlled, long-term study of
women who receive ei t her caf f ei ne or a placebo in order to see
clearly whether caf f ei ne has any impact on premenst rual symp-
toms, for such a st udy to be valid, however, the st udy subj ect s
would have to be ignorant about whether they were dri nki ng
caf f ei ne or not. This is hard to do because decaf f ei nat ed cof f ee,
tea, and cola generally t ast e di f f er ent from their caf f ei nat ed
versions. Perhaps more i mpor t ant , accordi ng to Rossignol, PMS
is still not wi del y considered a phenomenon with publ i c-heal t h
consequences, and so r ai si ng money for PMS research is ex-
t remel y di f f i cul t .
Tea for Two
Dur i ng a r out i ne pr enat al exami nat i on, a f et us being carried
by a young woman in the Net herl ands was f ound to have an
i rregul ar hear t beat . The woman was admi t t ed to the hospital,
where it was l earned that she dr ank more than a quar t and a
half of c a f f e i na t e d cola, two cups of coffee, and a mug of cocoa
every clay. The woman' s doctor advised her to abstain com-
pletely f r om t hese caf f ei ne- cont ai ni ng beverages. Within a
week, the f et us ' s hear t beat had r et ur ned to normal . The preg-
nancy progressed smoothly from t hen on (Grounds 1990).
This st ory serves two i mport ant purposes. Fi rst , it is a re-
mi nder t hat many women dr i nk caf f ei nat ed beverages duri ng
pregnancyt hough usual l y not to t hi s ext r eme. A 1983 study
of 1 , > 1 0 women98 percent of whom r egul ar l y consumed cof-
fee or tea pri or to becoming pr egnant f ound t hat 73 percent
continued to dri nk these beverages during pregnancy ( Kur ppa
ct al. 1983). Anot her st udy f ound caf f ei ne in the blood plasma
156 / Buzz
of 7S percent of the ncwborns t es t ed ( Dumas 1982). Second,
t he st ory is a r emi nder t hat c a f f e i ne , l i ke alcohol, f r eel y crosses
t he pl acent a. When a pregnant woman dr i nks a cup of cof f ee,
her unbor n chi l d exper i ences t he same degree of s t i mul at i on.
Women who nur s e t h e i r i n f a n t s need t o be car ef ul as well
because babies l ack the l i v er cn/ ymes needed to break down
c a f f ei ne. This impact i s pa r t i c ul a r l y i mpor t a nt f or babies born
pr emat ur el y. In one s t udy, t he h a l f - l i f e of c a f f ei ne i n pr emat ur e
i n f a n t s ranged f r om 41 t o 231 hour s. The average a dul t ha l f -
l i f e , in cont r as t , is S to 6 hour s .
The bi g quest i on, of course, i s whet her caf f ei ne r eal l y har ms
f e t u s e s or nur s i ng i n f a n t s . Ani mal s t udi es usi ng levels of caf-
f ei ne f ar hi gher t han any consumed by huma ns cl earl y dem-
onst r at e t hat c a f f ei ne can be a t er at ogent hat is, an agent
capable of caus i ng bi r t h def ect s . It was par t l y in response to
such s t udi es t hat in 1980 the l''DA advised pregnant women
t o r educe t hei r i nt ake of caf f ei ne t o a mi ni mum.
Si nce t hat t i me, ot her ani mal st udi es us i ng more t ypi cal caf -
f ei ne closes have shown an associ at i on between c a f f ei ne i nt ake
and lower bi r t h wei ght s, as well as increased i nci dences of still
b i r t h s and mi scar r i ages. A number of theories, none proved,
have been put f or war d t o expl ai n these ef f ect s ( Eskena/ i 1993).
For i ns t a nc e, c a f f ei ne doses as low as 200 milligrams (t he
a mount i n t wo cups of c of f ee) decreases placental blood fl ow.
Ca f f ei ne al so increases the f or ce of cont ract i on in the f et al
heart and decreases the levels of the most abundant ki nd of
estrogen: a r epr oduct i v e hormone called estradiol. Epi demi o-
logieal st udi es on h u ma n s have support ed t he vi ew t hat
cons umpt i on of moderat e to hi gh l ev el s of caf f ei ne duri ng
pregnancy (more t han 300 mi l l i gr ams a day) sl i ght l y i ncr eases
t he r i s k of spont aneous abort i on, i nt r aut er i ne growth r et ar da-
t i on (low bi r t h wei ght ) , and mi crocephal ya condition i n
whi ch t he head and br ai n arc abnor mal l y smal l .
The Body, Wired I 157
The evidence of developmental problems ari si ng f r om lower
levels of c a f f ei ne consumpt i on is less clear. Two studies, both
rigorously conduct ed and published in a leading medical j our -
nal, yielded very di f f er ent findings on the impact of low levels
of caf f ei ne consumpt i on. Although both f ound evidence for
i nt r aut er i ne growth r et ar dat i on among women who consumed
more t han 300 mi l l i gr ams of c a f f ei ne a day, one study (Mills
et al. 1993) found no ha r mf ul ef f ect s for doses lower than 300
milligrams, while the other ( I nf ant e- Ri v ar d et al. 1993) f ound
a si gni f i cant l y increased ri sk for spontaneous abort i on among
women who consumed doses as low as 163 mi l l i grams a day
duri ng t he f i r st t r i mest er .
Only a few st udi es have been conducted on children of
mot hers who dr ank eaf f ei nat ed beverages duri ng pregnancy.
One showed no ef f ect s of caf f ei ne on i nf ant and child neuro-
development ( Bar r and Streissguth 1991); another f ound
mat ernal caf f ei ne consumption to be responsible for poor
neuromuscul ar development and great er arousal and i r r i t abi l i t y
among babies (Jacobson et al. 1984). In light of these and many
other st udi es of f er i ng confl i ct i ng or di f f i cul t - t o- i nt er pr et dat a,
many doctors err on the side of caut i on and advise women who
are ei t her pr egnant or pl anni ng to conceive to abst ai n f r om
c a f f ei ne (Eskenazi 1993). The same advice is given to women
who br east - f eed, si nce caf f ei ne readi l y passes i nt o breast mi l k,
exposing an i nf a nt during a period of rapid neurodev el opment .
To the Bone
Some r ecent st udi es have given rise to popular concern about
caf f ei ne' s ef f ect s on bone densi t y and osteoporosis, a t hi nni ng
of the bones that t ends to occur l at er in l i f e, especially among
women. One such st udy resul t ed in headlines t hat , ironically,
may have exacerbated t hese f ear s even though the study resul t s
158/ Buz z
t hemsel ves did not directly implicate caf f ei ne as causi ng bone
loss.
The researchers st udi ed 980 women and f ound a small de-
crease in bone densi t y among women over the age of fifty who
had dr unk at least two cups of coffee a clay for many years
(Barret t -Connor et al . 1994). But the poi nt t hat got lost in
most medi a report s was t hat the loss was f ound only among
women who hadn't dr unk at least 8 ounces of milk a day on a
regular basis. Among those women who had consumed at least
a glass of milk a day, there was no increased risk of bone loss
even t hough t hey drank cof f ee. In other words, these resul t s
suggest that i t ' s not cof f ee dr i nki ng per se t hat seems to be
responsible for the loss of bone densi t y, but the reduct i on in
cal ci um i nt ake because some women dri nk cof f ee instead of
milk. The moral, then, is that everyone, whether they dri nk
coffee or not, should consume adequat e amount s of calcium.
Ot her st udi es have shown that nei t her the absorption of cal-
ci um from foods nor the excret i on of calcium from the body
is af f ect ed by caf f ei ne (Nil I 1994), leading most researchers to
conclude that caf f ei ne has no si gni fi cant impact on general
bone density or the disease of osteoporosis.
Decaf Ji tters
The glory years of decaf f ei nat ed cof f ee appear to be over. Back
in 1985, about 17 percent of the adul t coffee-dri nki ng popu-
l at i on drank decaf ( Nat i onal Cof f ee Association 1993). That
was up consi derabl y from the 4 percent figure in 1962. But
since 1985, fewer and fewer people t ake t hei r coffee "un-
leaded." By 1993 the percentage had dropped to 12.1 percent,
and the t rend shows no signs of hal t i ng.
The move away from decaf may have been spurred by resul t s
The Body, Wired I 159
from two separate st udi es hi nt i ng that decafbut not regular
coffeemi ght increase the risk of heart disease. The f i r st
st udy, conducted in 1989 at St anford University, showed that
in a group of 181 men, those drinking decaf experienced a 7
percent rise in their levels of low-density lipoproteins, the so-
called bad cholesterol. Although statistically significant, there
were reasons to doubt that the LDL rise was related to the
decaf. For i nst ance, the dat a didn' t show any relationship be-
tween the amount of decaf consumed and the degree of the
cholesterol rise. Finding such a dose response would have made
the case stronger. Even the director of the laboratory in which
the study was done said that he had no plans to change his
habit of drinking three cups of decaf a day (Lehman 1989).
A second study t racki ng the heal t h of 45,589 doctors for two
years f ound no association between caf f ei ne and heart dis-
easegood news for regular-coffee dri nkers (Grobbee ct al.
1990). But the data t urned up a "marginally significant" in-
crease in heart disease among the men who drank four or more
cups of decaf: a day. The study aut hor s i nt erpret ed their find-
ings cautiously. They noted that since the number of decaf
drinkers was relatively small, the resul t s were subject to larger
margi ns of error t han the data for regular coffee. They also
pointed out that t hey had st udi ed only men, and t hus their
findings might not apply to women. Finally, as with the Stan-
ford study, the ef f ect s found might have been at t ri but abl e to
some unexami ncd di f f er ence among the men in the two
groups.
Another popular concern about decaf is related to the cle-
caf f ei nat i on process i t sel f. Some f ear that substances used in
the process remai n in the beans and could pose a health t hreat .
Such f ear s are probably unwar r ant ed.
There are two basic ways to remove caf f ei ne from coffee
160 / Buzz
beans. In the "wat er process," green cof f ee beans are soaked
in hot wat er for ten mi nut es to two hours. In addi t i on to leech-
ing out most of t h e caf f ei ne, t hi s process removes most of the
compounds t hat give cof f ee i t s fl av or and body. Make cof f ee
f r om the beans at this stage, and a t r ul y wretched brew will
r esul t . That' s why ma nuf a c t ur er s t ake great pains t o r et ur n as
much of t hose lost flavor compounds as they can. Af t er the
soaking, t he caf f ei ne- l aced wat er is drawn off and the c a f f ei ne
is removed, ei t her wi t h a c a f f ei ne- s pec i f i c solvent ( such as
met hyl enc chl or i de or et hyl acet at e) or by passing the water
over aci d- t r eat ed carbon f i l t er s to whi ch the caf f ei ne binds. The
l i qui d is t hen r et ur ned to the beans, which rcabsorb some of
the fl avor compounds. After this step, the beans are dried and
s hi pped t o roast ers.
So-called sol v ent processing is more direct. Here, the green
cof f ee beans are washed wi t h a caf f ei ne solvent (agai n usual l y
met hyl ene chloride but somet i mes ethyl acet at e) in tubs or
r ot at i ng dr ums. The caf f ei ne i s t hen fi l t ered from t he l i qui d
sol v ent . Because this process is relatively f as t and because sol-
v ent s arc more speci fi c t han water in their action, more of t he
delicate flavor compounds arc usual l y ret ai ned in solvent-
processed beans.
Af t er the c a f f e i ne - l a de n solvent is removed, the beans are
processed with st eam to remove residual solvent. This removal
i s very ef f ect i v e because met hyl ene chloride and other c a f f ei ne
sol v ent s evaporat e at t emper at ur es between 100 and 120 de-
grees. In compari son, steam is 212 degrees and eof f ee beans
arc roast ed at t emper at ur es of 350 to 425 degrees. Vi r t ual l y no
solvent r emai ns by the t i me the cof f ee reaches your cup, which
is why the Food and Drug Admi ni st r at i on in 1985 ruled that
there is no risk in dr i nki ng solvent-processed decaf f ei nat ed
cof f ee.
The Body, Wired I I 6 1
The Last Drop
The ef f ect s of c a f f ei ne exami ned in t hi s chapt er are all acute
ef f ect s ; t hat i s, they t ake place shortly af t er i ngest i on. Whether
the ef f ect is heightened at hl et i c performance or an increased
urge to go to the bat hroom, these changes are all rel at i vel y
short-lived. Once the caf f ei ne has been metabolized, the ef f ect
di sappears.
But people tend to dri nk caf f ei ne on a regul ar basi s ewer long
periods of t i meof t en the great er part of a l i f et i me. What
ef f ect s might such long-term consumpt i on have on the body,
and how docs the brai n, in part i cul ar, respond to this si t uat i on?
In the section on headaches, we saw that blood vessels in the
brain qui ckl y adapt to caf f ei ne and that t hi s can lead to re-
bound headaches if caf f ei ne i nt ake is stopped abrupt l y. In the
next chapter, we' ll take a closer look at t hi s phenomenon as it
applies to the br ai n as a whole, and we'll look at the quest i on
of caffei ne' s addi ct i ve potential.
Caffeine Anonymous
"We were al l s t a ndi ng t her e t wi t chi ng, " Mi ke sai d, recount i ng
a t i me st andi ng in l i ne wai t i ng for cof f ee at a local caf e. "Ev-
eryone was sayi ng 'Come on, l et ' s go, l et ' s go. What ' s the
hol dup?' We were t i ke her oi n j unki es " (Ri chards 1995).
Mike is a member of the nat i on' s first chapt er of Ca f f e i n e
Anonymous, a suppor t group based in Port l and, Oregon, t hat
is modeled on Al cohol i cs Anonymous. Founded in April 1994,
the group r emai ns t i ny. Meet i ngs t ypi cal l y consist of between
Hooked I 163
five and eight people seated in a spare room in St. Stephen's
Episcopal Church. Size not wi t hst andi ng, members have strong
feelings about the substance t hat bedevils them: step 1 of the
group's twelve-step program calls on members to "admit that
we are powerless over caf f ei ne and that our lives have become
unmanageable. " For the group' s cofounder, Marsha Naegeli-
Moody, the word "addiction" exactly describes the compulsion
she felt at the height of her caf f ei ne consumpt i on. She was out
of control, she said, knocking back up to ten cups of coffee a
day. She predicts t hat others in the grip of caf f ei ne will f or m
similar groups in the f ut ur e to help those who want to kick
t hei r habi t . "In five years coffee is going to be t reat ed j us t like
nicotine, " she said in a newspaper interview.
Others find t hi s at t i t ude extreme. "Addicting? Hogwash,"
grumbled an on-line par t i ci pant in a 1995 Int ernet discussion
on the Portland group. "Habi t -formi ng, maybe. But let' s not
lump caf f ei ne in with heroi n, crack, and alcohol." At first
glance, t hi s kind of skepticism toward claims of caf f ei ne addic-
tion, or "caffei ni sm, " seems reasonable. In many ways, caf f ei ne
is in a di f f er ent league from other recreat i onal drugs. For one
thing, caf f ei ne' s power to i nt oxi cat e is relatively weak. The
buzz from a couple of cups of coffee is mild compared with a
typical hit of cocaine or amphetamine and is t ri v i al compared
with a dose of LSD.
The ef f ect of caf f ei ne is of t en so subtle t hat it is impossible
to tell if someone has consumed it or nota f act corroborated
by accident in one of the classic experi ment s on human reac-
tions to alcohol and caf f ei ne. In this study, subj ect s were given
alcohol, caf f ei ne, or a placebo beverage. An exami ner, who was
not told what the subj ect s had consumed, then tested the vol-
unt eers on everything from their reaction time to their emo-
tional state (Nash 1962). It t ur ned out that the exami ner eoulcl
i 64 / Buzz
easily t el l when a v o l u n t e e r had consumed alcohol, but the
exami ner could not tell when subj ect s had consumed caf f ei ne,
even t hough t he dose was si gni f i cant : 300 mi l l i gr ams.
The s ubt l et y of caf f ei ne' s ef f ect s is also ev i dent in the fact
t hat it is typically used to "normalize" rat her t han "intoxicate."
Caf f ei ne i s hel pf ul i n sust ai ni ng ment al f unc t i oni ng under con-
ditions of f a t i gue or boredom, such as t hat experienced by l at e-
s hi f t workers, st udent s crammi ng for an exam, or long-distance
t r uck dr i v er s. The onl y ot her recreat i onal dr ug used i n t hi s way
i s ni cot i ne, whi ch i s al s o seldom used for out ri ght i nt oxi cat i on.
Few would hes i t at e to f l y in a pl ane being piloted by a cof f ee
dr i nker or ci garet t e smoker, whereas an al cohol -guz z l i ng or co-
c a i ne - s nor t i ng pilot would be worrisome indeed.
Ca f f ei ne, of course, di f f e r s from ni cot i ne in t hat it is less
haz ar dous to one' s heal t h. In f act , caf f ei ne is arguably the saf-
est r ecr eat i onal drug. That, in any case, is the bot t om l i ne of
a great many s t udi es i nt o the mat t er . The ef f ect s of caf f ei ne
on such t hi ngs as breast cancer, bone loss, pancreat i c cancer,
colon cancer, heart di sease, l i v er di sease, ki dney di sease, and
ment al dys f unct i on have been exami ned i n exhaust i v e detail
and, to date, no clear ev i dence has been f ound l i nki ng mod-
erat e consumpt i on of caf f ei ne ( t he equi v al ent of three t o f our
cups of coffee dai l y) wi t h t hese or any other heal t h disorder
(Chou 1992; Gol ds t ei n 1994; Gorclis 1990; Grobbee et al.
1990). St i l l , c a f f e i n e i s not as pi r i n. Cert ai n i ndi v i dual s may be
unus ua l l y sensitive to caf f ei ne even wi t h low doses and may
exper i ence adv erse e f f e c i s such as increased anxi et y, or cardi ac
abnor mal i t i es such as pal pi t at i ons or heart ar r hyt hmi as . And,
as pr ev i ousl y noted, c a f f e i ne f r eel y passes t he pl aeent al bar r i er
and al so eas i l y ent ers breast mi l k, whi ch means t hat abstinence
is pr obabl y the safest choice for pr egnant or br east - f eedi ng
women.
Bal anced agai nst these speci fi c war ni ngs is t he f act t hat far
Hooked I 165
f r om killing people, caf f ei ne undoubt edl y saves livesthough
this is di f f i cul t to prove. St at i st i cs cannot be t abul at ed to de-
t er mi ne how many drivers might otherwise have fallen asleep
at the wheel had t hey not downed some coffee or a tablet or
two of No-Do/, before set t i ng out . This, of course, stands in
stark cont r ast to t he est i mat ed 400,000 Americans who die
each year from tobacco use, and the 100,000 whose deaths arc
at t ri but abl e to alcohol (Glass 1994).
And yet , despite its safet y and mildness rel at i ve to other
r ecr eat i onal drugs, caf f ei ne still unquest i onabl y al t ers brai n
f unct i on. These al t erat i ons trigger adaptive changes in the
brains of even casual users, r esul t i ng in such hal l marks of drug
addiction as tolerance, dependence, craving, drug-seeking be-
havior, and, af t er cessation, withdrawal sympt oms. As with al-
cohol, a mi nor i t y of users find caf f ei ne to be exceptionally
at t ract i v e; they crave it strongly, ingest ever larger amount s,
and s uf f er worse wi t hdrawal symptoms than do most users.
This is why the members of Caf f ei ne Anonymous cannot be
dismissed as overzealous handwringers, nor can their claim of
caffei ne' s addictive potential be swept aside as "hogwash."
"Caf f ei ni s m" is cert ai nl y a much less pressing societal prob-
lem than alcoholism or nicotine addi ct i on. It is highly unlikely
t hat membershi p in Caf f ei ne Anonymous will ever ri v al t hat
in AA. But caf f ei ne' s ability to induce drug reactions t hat re-
semble those experienced by addicts of t r ul y potent drugs is
widely unappr eci at ed.
Setpoint
The v ar i at i on in people' s physical responses to caf f ei ne is im-
pressive. Some people can dri nk several cups of coffee af t er
di nner, fall soundl y asleep an hour later, and sleep peacef ul l y
unt i l mor ni ng. Ot hers find t hat even one cup of coffee early
166 / Buzz
in the day i nduces a fitful ni ght ' s sleep. Likewise, the caf f ei ne
in a single cup of tea makes suscept i bl e i ndi v i dual s anxi ous
and unpl easant l y nerv ous, while for others c a f f e i ne is both a
r el axant and a mood-enhancer.
Some of t hi s v ari at i on is undoubt edl y rooted in genes. As
we've seen, the genes t hat give us uni que faces and f i nger pr i nt s
also give us uni que brai ns. No drug t hat acts on the brai n,
t herefore, is going to act exact l y the same way in everyone. No
one knows what kinds of i ndi v i dual di f f er ences lie at the bot-
tom of people' s divergent sensi t i v i t i es to c a f f ei ne. But there are
some hi nt s. For exampl e, people might d i f f e r in the number
and di st ri but i on of t hei r adenosine receptors. These receptors
arc manuf act ur ed accordi ng to bl uepri nt s stored in DNA. This
i nf or mat i on must be t r ans l at ed, t he ma nuf a c t ur i ng processes
carri ed out , and the f i ni shed receptors shepherded to t hei r
proper l ocat i ons in nerv e-cel l membr anes. All t hese steps re-
qui r e exqui si t el y del i cat e mol ecul ar controls, and v ar i at i ons i n
any of the steps could resul t in a person endi ng up wi t h more
or f ewer adenosine receptors t han nor mal . A person with an
above-average endowment of adenosi ne recept orst hat is,
someone with more t arget s for c a f f ei ne to hi t mi ght be hyper-
sensitive to c a f f ei ne. Conversely, people wi t h fcwer-t han-
normal adenosi ne receptors mi ght be unus ua l l y (^sensitive to
caf f ei ne.
Again, t hi s l i ne of reasoning is pur el y specul at i v e. No one
yet knows how much people v ar y in the quant i t y and quality
of t hei r adenosi ne receptors, nor is it completely clear what
ef f ect s such v ar i at i ons hav e in t er ms of behav i or. In real i t y,
genes probably af f ect c a f f ei ne sensi t i v i t y in doz ens of ways,
most of t hem not yet even guessed at. But genetic vari at i on
i sn' t the onl yand maybe not even the most i mpor t ant r ea-
son people d i f f e r so much in t hei r react i ons to caf f ei ne (Dews
1984). Al t hough it may sound l i ke ci rcul ar r easoni ng, one of
Hooked I 167
t he reasons people di f f e r i n their response t o c a f f ei ne i s t h a t
people di f f e r in t hei r consumption of caf f ei ne. Consumpt i on,
in other words, can radically af f ect sensitivity.
Many habitual drinkers of caffei ne-cont ai ni ng beverages find
that they must increase t hei r dose to achieve the preferred
degree of st i mul at i on. Lying behind this phenomenon is t he
brain' s remarkable pl ast i ci t y. To an ext ent far great er t han any
other organ, the brai n adapt s to ehanging eondi t i ons. It has to.
In addition to being the seat of consciousness and awareness,
the brain controls heart beat , breat hi ng, and ot her l i f e- suppor t
systems. Wild f l uct uat i ons in brain activity owing to changing
env i ronment al conditions would t hus put the rest of the body
at severe risk. Shaped by millions of year s of such selective
pressure, the human brain today comes equi pped with dozens
of mechani sms designed to t i ght l y regulate the level of br ai n
act i v i t y. Like thermostats, they const ant l y a dj us t such t hi ngs
as neur ot r ans mi t t er release and receptor sensi t i v i t y to compen-
sate for pert urbat i ons from the env i ronment .
The si t uat i on is analogous to another regul at or in the
bodythe one controlling wei ght . As most di et ers know from
hard experience, the body has a "setpoint"a weight t hat it
strives to mai nt ai n despite fluctuations in food i nt ake. This
setpoint v ari es between i ndi vi dual s and is f undament al l y gov-
erned by genes (Leibel et al. 1995). Research has shown t hat
when calories are cut, metabolism slows in compensation. If
excess calories are consumed, metabolism speeds up in an ef f or t
to bum off the extra calories and bri ng body weight back to
the setpoint.
The same principle applies to the brai n' s setpoint. Attempts
to rev up brain activity ( f or instance, with c a f f e i ne ) are met
with a countervailing response that reduces brai n act i v i t y. At-
tempts to lower brain act i v i t y (as with alcohol) are met wi t h
the opposite response. The brai n constantly st ri ves to mai nt ai n
168/Buzz
its genetically governed setpoint of activity, even if t hi s means
going to ext r aor di nar y ends to achieve it. This flexible response
to any drug, whet her recreat i onal or t her apeut i c, is called tol-
erance. Longtime heroin users, for i nst ance, have been ob-
served to r equi r e ten thousand times the dose t hey i nj ect ed
when t hey began t hei r habi t . Their br ai ns adapt to heroi n to
such an extent t hat they i nj ect themselves with quant i t i es of
t hi s narcotic t hat could kill a person not t ol er ant to heroi n. In
cont rast , the tolerance achievable by even the heaviest cof f ee
dri nker rarel y requi res more than ten to f i f t een times the caf-
f ei ne a first-time dri nker might consume (Goldstein 1994).
That hardly means, however, t hat tolerance is a trivial i ssue.
Up Escalator
Research with drugs such as heroin has revealed that one way
the brain responds to drug-induced per t ur bat i ons is to change
the number of receptors in the af f ect ed neur ot r ansmi t t cr sys-
tem. This now appears to be one way the brai n reacts to caf-
fei ne.
As we' ve seen, caf f ei ne plugs adenosine recept ors, t hus
blocking their normal ability to slow the brain clown. This
blockage is detected via an unknown mechani sm and triggers
the creation of more adenosine receptors. It' s as though the
receptors were ant ennae pi cki ng up a st eady r adi o signal; when
the signal suddenl y weakens, more ant ennae are added to the
system to compensate. This adapt i v e process of i ncreasi ng
receptors is called up-regul at i on, and it is a common brain
response to any drug that blocks a specific circuit or neuro-
t ransmi t t er. The opposite response, called down-regul at i on is
t ypi cal l y seen in reaction to drugssuch as her oi nt hat di-
rectly stimulate neur ot r ansmi t t er receptors.
A number of st udi es have shown t hat caf f ei ne and ot her
Hooked I 169
methylxanthines up-regulate adenosine receptors in many tis-
sues, including the brain (Fastbom and Fredholm 1990; San-
ders and Murray 1988). Up-regulation t hus may be responsible
at least in part for tolerance to caf f ei ne, though, again, how
this happens is not clear, and studies on whet her caf f ei ne
causes adenosine-rcceptor up-regulation have yielded contra-
dictory results ( Kapl an et al. 1993; Zielke and Zielke 1987).
It appears t hat chronic caf f ei ne use may cause up-regul at i on
or down-regulation of ot her neur ot r ansmi t t cr systems as well.
Recent experiments with mice revealed the expected 20 per-
cent up-regulation of A, adenosine receptors, but the scientists
also found sur pr i si ng changes in receptor densities for many
other important neurot ransmi t t ers (Shi et al. 1993). Receptors
for norepi nephri ne (a hormone similar to adrenal i ne) were re-
duced. Densities of cert ai n serotonin receptors were increased,
as were densities of acetylcholine receptors. And a striking 65
percent up-regul at i on of GABA receptors was observed. These
results suggest that caf f ei ne i ndi rect l y af f ect s many neurot rans-
mi t t er systems through its direct ef f ect s on adenosine recep-
tors. The added firing in many brain ci rcui t s owing to caf f ei ne
intake undoubt el y increases or decreases the release of clopa-
mine, serotonin, and other i mpor t ant neur ot r ansmi t t er s. It is
too early to say what ef f ect these secondary al t erat i ons have on
behavior, t hough they may t ur n out to be i mport ant compo-
nents of the general experience of being "wired" on caf f ei ne.
Tolerance to caf f ei ne sets in rel at i vel y qui ckl y. Animal ex-
peri ment s wi t h large doses of caf f ei ne have induced tolerance
in as little as three days (Dal y 1993). Tolerance in humans
develops a bit more slowly, probably because humans do not
ingest the large amount of caf f ei ne typically admi ni st ered to
test ani mal s. Still, humans generally become tolerant to a given
dose of caf f ei newhet her a single can of soda or ten cups of
cof f eei n a week to twelve days (Regestein 1995).
170 / Buzz
This tolerance can be remarkabl y complete; that is, the
brai n' s a bi l i t y to compensat e for caf f ei ne can be so ef f ect i v e
t hat t ol er ant users experi ence very l i t t l e, i f any, t rue st i mu-
l at i on by t hei r cust omar y close. This was demonst r at ed by a
par t i cul ar l y rigorous exper i ment i n whi ch t hi rt y-t wo heal t hy
vol unt eers part i ci pat ed in a mont hl ong study of the subject i ve
ef f ect s of caf f ei ne (Evans and Gr i f f i t hs 1992). Half the volun-
teers received c a f f ei ne (in capsule f or m) , and half received a
placebo. The people consumi ng the caf f ei ne demonst rat ed tol-
erance in several ways. On so-called forced-exposure clays in
which all par t i ci pant s received caf f ei ne, those who had been
t aki ng the placebo showed much great er ef f ect s from the caf-
fei ne t han t hose who had been t aki ng c a f f ei ne regul arl y. The
scores for the placebo group on such t hi ngs as t ensi on, anxi et y,
j i t t er i nes s , and t he perceived st r engt h of t he dr ug were oft en
several t i mes greater t han t he scores of those in the caf f ei ne-
tolerant group.
Most t el l i ng were observ at i ons made dur i ng t he portion of
the experi ment in which ingestion of ei t her a placebo or caf-
fei ne was held const ant for ei ght een days. The two groups were
given a bat t er y of t est s to rate everyt hi ng f r om their mood to
t hei r physical heal t h. Remarkably, even though the people in
the c a f f e i n e group were consumi ng 900 mi l l i grams a day, the
average scores for bot h groups were v i r t ual l y i dent i cal . The par-
t i ci pant s r epor t ed no si gni f i cant di f f er ences in such t hi ngs as
energy, al ert ness, i r r i t abi l i t y, t al kat i v eness, t ensi on, and depres-
sion/ dejection. Evi dent l y, the br ai ns of the caf f ei ne consumers
had adapted f ul l y and relatively qui ckl y t o caf f ei net o t he
ext ent that t hey were "normal," at least compared with those
of t hei r non- caf f ci ne- consumi ng peers.
This raises an obvious quest i on: If people become t ol erant
to caf f ei ne in a mat t er of days and t her eaf t er deri ve es s ent i al l y
no st i mul at i on from the drug, what account s for the enormous
Hooked I 171
popul ar i t y of caf f ei ne- cont ai ni ng beverages (other than the fact
that many people find them delicious) and for the distinct
sense by users t hat t hey are, in f act , being st i mul at ed? One
possibility is that some part s of the brai n may not become
tolerant to caf f ei ne. Even heavy long-term users may t hus be
feeling some kind of "buzz," through the general muf f l er of
tolerance (Nehlig et al . 1992). Evidence supporting this idea
is, however, still qui t e tenuous, and even if it' s t rue, the ef f ect
is likely to be r at her subtle. A more likely explanation for why
people cont i nue to consume caf f ei nat ed beverages long a f t er
tolerance has been established can be found by looking at the
flip side of the phenomenon of tolerance: wi t hdrawal .
On the Rebound
Since caf f ei ne is rel at i v el y i nexpensi v e and widely available, the
dose escalation i nduced by tolerance is seldom burdensome.
Nobody must resort to crime to support their habit, nor do
they need to rely on back-alley dealers to supply their daily
fixes. Of course, tolerance to very heavy doses of caf f ei ne may
be problematic for health reasons. Both coffee and tea, for in-
stance, are f ai r l y acidic beverages, and some people find t hat
ingestion of large amount s i rri t at es their stomach. But by and
large, tolerance isn't even noticed as long as circulating levels
of caf f ei ne are kept stable. The problems come when those
levels drop, at which time the brain' s delicately balanced see-
saw of neur ot r ansmi t t er s and receptors tips radically. Without
the "weight" of caf f ei ne to push against, the br ai n goes over-
board. The result is wi t hdr awal : a constellation of physical and
psychological symptoms that in the ease of caf f ei ne can range
from imperceptible to intensely unpl easant t hough caf f ei ne
withdrawal is never lethal the way wi t hdrawal from alcohol or
heroin can be.
172 / Buzz
By far the most common symptom of caf f ei ne wi t hdrawal is
headachea f act t hat has only recent l y been aeeepted by the
medical communi t y. This accept ance grew out of the long-
observed phenomenon t hat many pat i ent s given general anes-
t hesi a exper i ence a headache when t hey come to. Such
post operat i v e headaches have t r adi t i onal l y been considered an
unav oi dabl e side ef f ec t of the anest hesi a i t s el f . But in 1989,
three doctors at Hammer s mi t h Hospi t al near London ques-
tioned t hi s assumpt i on (Galletly ct al . 1989). They decided to
test a not her t heory about the origin of the post operat i v e head-
ache: t hat it is due to c a f f e i ne wi t hdr awal i ni t i at ed by the stan-
dard r equi r ement t hat pat i ent s under goi ng elective surgery
involving general anest hesi a abstain from both food and caf-
f ei nat ed beverages pri or to t hei r operat i on.
The doct ors i nv est i gat ed t hei r hunch by hav i ng 142 ran-
domly sel ect ed pat i ent s fill out a quest i onnai r e a f t er they re-
covered f r om their anest hesi a. The surv ey asked how the
pat i ent s f el t as well as what t hei r t ypi cal i nt ake of caf f ei ne was
before the sur ger y. The r es ul t s showed t hat the doctors were
on t o somet hi ng: t he more c a f f e i ne pat i ent s consumed, t he
more l i kel y t he y were t o experi ence headache af t er anest hesi a.
Seventy-three percent of the pat i ent s consumi ng more t han
100 mi l l i gr ams of caf f ei ne a day experienced headache; at the
other end of the spect rum, none of the six pat i ent s who con-
sumed no c a f f ei ne pri or to surgery had a headache a f t e r anes-
t hesi a.
That it l i a s t aken so long for t hi s seemi ngl y st rai ght forward
associ at i on between caf f ei ne i nt ake and post operat i ve head-
ache to be accepted says v ol umes for the popul ar perception
of c a f f ei ne. U nt i l r ecent l y, c a f f e i n e was not considered a drug
by ci t her the general public or doctors, hence the idea t hat
sudden cessation of caf f ei ne could preci pi t at e wi t hdr awal
sympt oms wasn' t consi dered.
Hooked I 173
Since that 1989 study, several other maj or investigations
have unequi vocal l y demonst rat ed the reality of caf f ei ne with-
drawal symptoms (Silverman et al. 1992, Strain et al. 1994).
Here are the typical symptoms of caf f ei ne wi t hdrawal in rough
order of t hei r occurrence in the general population:
Headache
Depression
Fatigue
Lethargy
Irritablcness
Increased muscle tension
Nausea
Vomiting
Silverman' s st udy included some direct quotes from people
experiencing caf f ei ne withdrawal:
"I fel t like I had the flu, a severe headache, extreme fatigue. "
"I felt sad, uncert ai n about the f ut ur e, a general feeling of
glum."
"I couldn' t concentrate even when I had to do those t est s.
I'm basically not a low person: I was mildly sad and depressed."
The most extreme response came from a woman who said,
"I had a severe headache that progressed into vomiting,
flu-like symptoms. I can only compare that sickness to the
radi at i on and t reat ment (radi at i on and chemotherapy t reat -
ment for cervical cancer) of the past year. It was as bad as
that. "
Not surpri si ngl y, the st udi es of caf f ei ne tolerance and with-
drawal have f ound wide v ari at i ons in subject responses. Some
people, even heavy consumers, report no withdrawal sympt oms
at all, while others s uf f er severe headaches and other unpleas-
ant symptoms even though they consumed as little as one cup
174 / Buzz
of cof f ee a day pri or to abst ent i on. In general, wi t hdrawal
sympt oms begin wi t hi n twelve to t went y- f our hour s af t er the
last use and peak anywhere from t went y to fort y-ei ght hours
a f t er caf f ei ne consumpt i on stops (Hughes 1992). Withdrawal
symptoms then t ypi cal l y t aper of f , but it usual l y takes a f ul l
week for a r et ur n to normal .
These timings are si gni f i cant . Most regul ar consumers of caf-
fei ne are in the first stages of caf f ei ne wi t hdrawal when they
wake up in t he morni ng ( assumi ng they didn' t dr i nk coffee j us t
before going to bed the previous ni ght ). For t hi s reason, caf -
f e i ne users, in general , are likely to feel more tired, irritable,
and groggy in the mor ni ng t han people who abst ai n from caf-
feine. Also, if mor ni ng caffei ne i nt ake is skipped, a headache
is likely l at er in the morni ng or early t hat af t er noon. Dri nki ng
c a f f ei ne, of course, qui ckl y al l evi at es these wi t hdrawal symp-
toms, j us t as the classic "hair-of-the-dog" nip of alcohol
"cures" a hangov er. It ' s not surpri si ng, t herefore, that the
mor ni ng r i t ual for many caf f ei ne consumers involves getting
caf f ei ne into the bloodstream as qui ckl y as possible. The pres-
ence of morni ng wi t hdrawal symptoms also explains why the
first cup of coffee, tea, or soda can give the most pronounced
boost to mood and energy: the ef f ect is more obvious because
of t he cont rast with the feel i ngs of lethargy and depression
associated with wi t hdrawal .
Thanks in part to the st udi es j us t ci t ed, the r eal i t y of caf f ei ne
wi t hdr awal is coming to be appreci at ed by both physicians and
the lay publ i c. Physicians, for i nst ance, have been advised to
ask about caf f ei ne use when pat i ent s complain of symptoms
such as headaches, depression, f at i gue, and drowsiness (Hughes
1992). Another consequence of the growing appreciation for
caf f ei ne withdrawal is the recogni t i on t hat , for a mi nor i t y of
users, caf f ei ne may have addi ct i v e potential.
Hooked I 175
Dependence
Tolerance and withdrawal are the classic signs of physical de-
pendence on a drug. But physical dependence isn't the same
thing as addiction. Morphine and Valium, for example, regu-
larly produce physical dependence. The brain down-regulates
the opiate and GABA receptors that are the immediate targets
of these two drugs, resul t i ng in a need for larger and larger
doses as time goes on. But opiates and the benzodiazepines
have i mport ant medical uses in the management of pain and
anxi et y. Their controlled use by patients who have been care-
f ul l y screened, well i nformed, and closely monitored by physi-
cians does not constitute addiction even though such patients
clearly become physically dependent on the drug a f t er a period
of regul ar use.
So what does const i t ut e addiction? The arbi t er of such things
these days is the latest version of the Diagnostic and Statistical
Manual of Mental Disorders (DSM-IV) of the Ameri can Psy-
chiatric Association. The DSM-IV defines two kinds of problem
relationships with drugs: substance abuse and the more serious
substance dependence. This latter category is what most people
would call addiction, though the DSM-IV avoids that word
because it is so heavily freighted with moral and emotional
connotations.
Substance dependence is characteriz ed by using a substance
in larger amount s or for a longer period of time t han intended;
repeated unsuccessful ef f or t s to cut down or control use; use
of a substance to relieve or avoid withdrawal symptoms; or a
pat t ern of compulsive drug-taking that persists despite clear
social, psychological, physical, or occupational problems related
to the drug. For decades, no one thought that caf f ei ne was a
potent enough drug to cause the serious problems associated
176 / Buzz
with subst ance dependence. But t hat view has been revised in
light of a recent st udy by Eric St rai n and his colleagues at the
Johns Hopki ns Uni v er si t y School of Medicine. The researchers
wanted to see whet her they could find caf f ei ne users who
woul d q ua l i f y as substance dependent under t he DSM-IV def -
i ni t i on ( St r ai n ct al. 1994). Newspaper ads were used to locate
people who t hought they were psychologically or physically de-
pendent on caf f ei ne. Out of ni net y- ni ne people screened for
t he st udy, si xt een were diagnosed as caf f ei ne dependent a f t er
undergoi ng a bat t ery of ev al uat i ons.
The average daily consumpt i on of caf f ei ne of these part i ci -
pant s was 357 milligrams, which is somewhat but not st ri ki ngl y
higher t han the 280-milligram average consumed in the United
States. The act ual daily amount consumed by t he i ndi v i dual s
ranged from a low of 129 mi l l i grams to a high of 2,548 mi l l i -
grams. Half the subj ect s got t hei r caf f ei ne from cof f ee, 44 per-
cent drank caf f ei ne- cont ai ni ng soda, and one part i ci pant drank
t ea. Fully 81 percent of the caf f ei ne- dependent subj ect s said
they had made unsuccessf ul ef f or t s to cut down or control their
use, and 94 percent said they consumed caf f ei ne despite a per-
sistent or recurrent physical or psychological problem related
to its use. Almost hal f of the subjects reported some physical
condi t i on such as heart pal pi t at i ons or gast roi nt est i nal prob-
lems t hat had led t hei r physicians to recommend reducing or
el i mi nat i ng c a f f ei ne consumpt i onand in each case the sub-
j ect s were unable to do so.
In a second phase of t hi s st udy, the researchers tested eleven
of t he si xt een caf f ei ne- dependent subj ect s for wi t hdrawal
symptoms. I nt er es t i ngl y, two of the eleven showed no with-
drawal sympt oms at all, though they told the researchers
t hat they had experienced such sympt oms when they tried to
qui t in the past . This finding is f ur t her evidence of the wide
Hooked I 177
v ar i abi l i t y i n wi t hdrawal ef f ec t s f r om person t o person and
even among i ndi v i dual s t hemsel v es f r om one episode to the
next .
A v ar i et y of i mpai r ment s were report ed by subject s when
they were in wi t hdrawal , i ncl udi ng
Missing work owing to bout s of v omi t i ng
Maki ng mul t i pl e costly mi st akes at work
Going home f r om work early to sleep
Inabi l i t y to complete schoolwork
Screaming at the children
Cancellation of a son's bi r t hday par t y
Being too t i red to do househol d chores
In addition to under scor i ng the seri ousness of caf f ei ne with-
drawal and set t l i ng the quest i on of whet her some people really
do qual i f y as "caf f ei ne addicts," the Johns Hopkins researchers
uncovered an i nt r i gui ng pat t er n in the dat a. Fifty-seven per-
cent of those diagnosed as caf f ei ne dependent had earlier been
diagnosed as s uf f er i ng f r om ei t her alcohol abuse or alcohol de-
pendence. In addi t i on, seven of the si xt een subj ect s had a past
diagnosis of manic-depressive disorder or maj or depression.
The researchers suggested t hat this cl ust eri ng of caf f ei ne de-
pendence wi t h alcohol abuse and/or mood di st ur bances de-
serves f u r t h e r st udy. Do some people t ur n to caf f ei ne and other
drugs because they are t ryi ng to sel f-medi cat e an underl yi ng
mental dys f unct i on? Does caf f ei ne use by such people some-
how exacerbat e or i ni t i at e ot her substance-abuse problems? Is
t here an "addictive personal i t y" predisposed to dependence on
many types of drugs? These are j us t some of many quest i ons
awaiting a f ul l er under s t andi ng of the biological and psycho-
logical under pi nni ngs of addiction in general.
178 / Buzz
Caffei ne Paradox
Consi der the following r eal - l i f e case studies:
A t hi rt y-fi v e-year-ol d of f i ce worker who sleeps twelve
hour s a ni ght , f al l s asleep every time she watches tele-
v i si on, and st ays in bed all day on Sunday, even t hough
she dr i nks ten cups of cof f ee and two liters of cola a day
A fifty-two-year-old secr et ar y who oversleeps regul arl y
and feel s unbearabl y groggy i n mi daf t er noon despite her
dai l y consumpt i on of six or seven cups of coffee as well
as a prescribed s t i mul a nt
A f or t y- f i v e- year - ol d cabi net maker who wakes up groggy,
takes a nap every day, f al l s asleep over meals, and has
out bur st s of temper, even t hough he dri nks six or seven
cups of coffee and day and suppl ement s t hem with caf-
f ei ne pills.
In each of these cases the disagreeable sympt oms disappeared
when t he pat i ent s stopped t aki ng caf f ei ne or other st i mul ant s
( Caf f ei ne 1990). For these people, in other words, caf f ei ne was
act i ng as a depressant , not a s t i mul ant : i nst ead of energy, mo-
t i v at i on, and hei ght ened mood, t hey experi enced l et har gy,
sl eepi ness, and depression.
There are several possible expl anat i ons for t hi s paradoxical
ef f ec t of caf f ei ne, according to Ouent i n Regest ei n, the director
of the Sleep Clinic .of Brigham and Women' s Hospi t al in Bos-
t on, where these pat i ent s were t reat ed. For instance, the heavy
c a f f ei ne use by t hese people might have i nt er f er ed with their
sleep at ni ght to such an extent that they were simply ex-
haus t ed dur i ng t he dayso much so that more caf f ei ne could
not overcome t hei r t or por . Or per haps these people were sim-
Hooked I 179
ply hypersensi t i v e to the depressant ef f ect s of v er y high doses
of caf f ei ne, Regestein believes that the answer lies buried in
the complicated v ar i at i ons of receptor profi l es and ncur ot r ans-
mi t t ers among i ndi v i dual s. "This is why medi ci ne isn' t a sci-
ence," he says. "We j ust don' t know what' s going on with these
people. All we know is t hat they i mprove when caf f ei ne use is
stopped."
Another example of a paradoxical ef f ect , si mi l ar to the one
observed by Regestein, involves the use of st i mul ant s in the
t reat ment of at t ent i on deficit hyperact i v i t y disorder ( ACI I D).
The ser endi pi t ous discovery in the 1930s of the calming i nf l u-
ence of cert ai n st i mul ant s on the behavi or of some children
diagnosed as hyper act i v e led to a search for al t ernat i v e st i mu-
l ant s that would be ef f ect i v e, non-habit f or mi ng, and i nexpen-
sive (Gi t t el man 1983). Ca f f ei ne was one of the first such drugs
to be studied.
Of the seven controlled st udi es of caf f ei ne and hyper act i v i t y,
five fai l ed to detect any advant age of caf f ei ne over a placebo;
two reported si gni fi cant i mprovement . The positive resul t s of
these l at t er two studies have been quest i oned beeausc of t hei r
rel at i v el y small sample si/e. Since no st udy reported t hat caf-
f ei ne was worse t han a placebo, a consensus emerged t hat
caf f ei ne probably has a weak but clinically uns a t i s f a c t or y ther-
apeut i c i mpact on chi l dren labeled as havi ng ADHD.
The neurological mechani sms behind the ef f ect s of st i mu-
l ant s such as caf f ei ne or Ri t al i n (a popular drug t r eat ment for
ADHD) are not yet known. From a t heoret i cal point of view,
however, t hi s ki nd of paradoxical ef f ect isn' t i nherent l y mys-
t er i ous. Many of the neurons in the cerebral cort ext he seat
of "higher" f unct i ons such as rat i onal thought, speech, and
creat i v i t yinhibit the f unct i oni ng of other par t s of the brai n.
When such neur ons fire, t hey dampen act i v i t y el sewhere, which
is, appar ent l y, a cri t i cal f unct i on in a heal t hy brain. Increasing
S 80 / Buzz
the f i r i ng rat e of t hese cortical neur ons by s t i mul ant s , in other
words, could have the par adoxi cal ef f ect of t ur ni ng down the
"volume" in ot her br ai n ci rcui t s, perhaps allowing for the in-
creased at t ent i on span and abi l i t y to concent rat e t hat is some-
t i mes observed when hyper act i v e chi l dr en are given st i mul ant s.
Since most people do not experi ence a depression of cortical
f unc t i on when they take s t i mul ant s such as caf f ei ne, t he kind
of paradoxi cal i nhi bi t i on seen in some sleep-disorder pat i ent s
and hyper act i v e chi l dr en may be due to an under l yi ng neur o-
logical di f f er enc e i n t hei r br ai n chemi st r y. Much cur r ent re-
search is aimed at t est i ng t hi s idea in the hope of finding more
ef f ect i v e t r eat ment s for both problems.
Vari ati on Redux
In t hi s chapt er, we've seen that the brain qui ckl y adapts to
caf f ei neas it docs to other dr ugsbecause it cont i nual l y
st ri v es to mai nt ai n a "sct poi nt " of neurological act i v i t y. Such
a da pt a t i ons l i e behind the phenomenon of tolerance, and tol-
erance expl ai ns why regul ar user s of c a f f ei ne experi ence a re-
duced "kick" f r om t hei r st andar d dose wi t hi n a mat t er of days.
Tolerance al so expl ai ns why many people experience with-
dr awal sympt oms when t hey stop dri nki ng caf f ei neat ed bever-
ages. Having adapt ed to caf f ei ne, the brai n "rebounds" in its
absence, produci ng a const el l at i on of unpl easant sympt oms
such as headache, f at i gue, depressi on, and i r r i t abi l i t y.
Tolerance and wi t hdr awal are signs of physical dependence,
which is nut the same as ei t her substance abuse or substance
dependence ( addi ct i on) . Although most caf f ei ne user s are
physically dependent to one degree or anot her, many addi t i onal
fact ors must be weighed before the labels of abuse or addiction
can be used. Only for a mi nor i t y of users can c a f f ei ne be
termed "addictive. "
I looked I 181
We end our expl orat i on of caf f ei ne, t her ef or e, exactly where
we ended our look at alcohol: contemplating the range of hu-
man biological di v ersi t y. The practical consequence of t hi s di-
v er si t y is the nonexistence of blanket rules or guidelines for
caf f ei ne use. In the end, the best answers come from personal
experi ment at i on with v aryi ng doses of caf f ei ne to sec how this
drug i nt er act s with one's uni que biochemistry. As with alcohol,
the i nf or mat i on presented here about how caf f ei ne works is
probably most hel pf ul as a baseline against which to gauge
one' s personal experiences.
The Missing Li nk
Up to t i ns poi nt we' ve i gnor ed a rat her i mpor t ant fact about
alcohol and c a f f ei ne: t hat many people end up with both drugs
ci r cul at i ng t hr ough t h e i r brai n at t he same t i me. This some-
t i mes r esul t s f r om the consumpt i on of a dual - dr ug beverage
such as I r i s h cof f ee (whiskey and cof f ee) or rum and Coke.
More commonl y, beverages cont ai ni ng alcohol or caf f ei ne are
consumed in close t emporal pr oxi mi t y to one anot her , as when
a meal begins wi t h wine and ends wi t h espresso, or when a
mar t i ni follows a long day of slugging cof f ee at the of f i ce.
What happens u n d e r t hese ci r cumst ances? How do alcohol
Better Living Through Chemistry I 183
and caf f ei ne i nt eract ? Even though t hey are opposites in many
ways, they clearly don't simply anni hi l at e each other on contact
like mat t er and ant i mat t er . People who consume f our Irish cof-
fees in rapid succession are anything but sober. But they will
not be feeling either purel y intoxicated or purel y wired. What
manner of i nebr i at i on will they be experiencing, and what neu-
robiology supports it?
For a long time, it was believed that caf f ei ne and alcohol
went their separate ways in the br ai n. It was t hought that t hey
worked on f undament al l y d i f f e r e n t brai n circuits and neuro-
t ransrni t t er syst ems and that they did not, t her ef or e, directly
antagonize each other' s actionsan idea enshri ned in the stan-
dard advice that if you try to sober up a dr unk with caf f ei ne
you' ll simply end up with a wide-awake dr unk. This assumpt i on
is sound, even though the premise on which it's based has been
proved wrong.
Research has shown that there is a direct link between the
actions of alcohol and caf f ei ne. The two drugs count eract each
other' s influence on one of the brai n' s i mport ant neurot rans-
mi t t er systems, which means t hat , to a limited extent anyway,
caf f ei ne and alcohol can neut ral i z e each other. The first hi nt s
of this rel at i onshi p arose f r om alcohol st udi es usi ng the mice
we met in Chapter 5: long-sleep mice, which become comatose
on low doses of alcohol, and short-sleep mice, which tolerate
relatively high doses and nap only bri efl y when finally over-
come.
In the early 1980s, neur osci ent i st s were t ryi ng to pin down
the neurological basis for these markedly di f f er ent reactions to
alcohol. It proved to be qui t e di f f i cul t . Neur ot r ansmi t t er sys-
tem a f t er neur ot r ansmi t t er system was examined and f ound to
be essentially identical between the two st rai ns of mice. Finally,
in 1984, a significant neur ot r ansmi t t er di f f er ence was f ound in
the then relatively obscure adenosine system.
William Proctor and Thomas Dumviddie in the Depar t ment
184 / Buzz
of Pharmacol ogy at the Uni v er s i t y of Colorado' s Heal t h Sci-
ences Cent er discovered t hat shor t - and l ong-sl eep mice re-
sponded v ery d i f f e r e n t l y t o drugs af f ect i ng adenosi ne. For
i nst ance, a drug called L-PIA, which mimics adenosine, caused
the l ong- sl eep mice to become very sleepy and l et hargi c, while
hav i ng l i t t l e ef f ect on t he shor t - sl eep mice (Proctor and Dun-
wi ddi e 1984). When the r es ear cher s gave the mice theophyl-
l i ne , which ant agoni z es adenosi ne recept ors, t he long-sleep
mice were 61 per cent more active t han usual . The short-sleep
mice, i n cont r ast , showed no i ncrease i n act i v i t y a f t e r t he i n-
j ec t i on. In short , t hese two s t r ai ns of mice, which react very
di f f e r e nt l y t o alcohol, al so reacted v ery d i f f e r e n t l y t o drugs af-
fect i ng adenosi ne. These were st ri ki ng results because they im-
plied a st rong neur ochemi cal connect i on between t he two most
popul ar dr ugs on t he pl anet . Despi t e i t s i mpl i cat i ons for every-
day consumer s of alcohol and c a f f ei ne, however, the st udy find-
ings di dn' t make headl i nes. This f undament al rel at i onshi p
between t he act i ons of alcohol and c a f f ei ne has t hus remai ned
v i r t ual l y unknown to all but a few ncur osci ent i st s who special-
ize in adenosi ne.
Among t hose sci ent i st s, however, t he paper set off a search
for the mol ecul ar mechani sms under l yi ng t he obser v at i ons i n
mice. The obvious place to start was to see whet her alcohol
di r ect l y af f ect s adenosi ne levels i n t he br ai n. Ini t i al report s
have been posi t i v e: when neurons are exposed to alcohol, aden-
osi ne levels i ncrease i n t hei r v i ci ni t y. Si nce adenosi ne of t en
depresses nenr onal firing, its l i ber at i on by alcohol woul d con-
t r i but e to the sedat i on and l et har gy experi enced by people who
dr i nk moder at e to heav y doses.
Kx a c i l y how alcohol triggers adenosi ne release is not yet un-
derst ood. One pr omi si ng i dea is t hat al cohol disables a molec-
ul a r pump t hat nor mal l y sucks up f r ee adenosi ne and
t r anspor t s it back into the cell i nt er i or (Gordon et al. 1993).
Better Living Through Chemistry I 185
Alcohol appears to di sr upt t hi s adenosine pump, j us t as it in-
t er f er es with so many of the brai n' s other f unct i ons . The im-
pai rment of t hi s key t r anspor t er could leave excess adenosi ne
outside nerve cells, t hus expl ai ni ng the actions mentioned
above.
Still, even though the t ransport er theory seems sound, it is
too early to say with confidence that t hi s is, in f act , the long-
sought l i nk between alcohol and caf f ei ne ( Dunwi ddi e 1995).
Too l i t t l e is known about how adenosi ne pumps work and
where t hey are located. Given alcohol' s wide-ranging ef f ect s ,
the mechanism behind the observed bui l dup of adenosine
could lie someplace else ent i rel y.
Regardless of how alcohol and adenosine are connected,
there is l i t t l e doubt t hat the connection exists. And t hat raises
an obvious quest i on: If alcohol i nt oxi cat i on involves increased
adenosine levels, shouldn' t caf f ei ne count eract drunkenness?
Antagonism
Caf f ei ne and alcohol have been used as ant i dot es for each
other for cent ur i es. In the earl y years of coffee' s i nt roduct i on
to Europe, for instance, the French wr i t er Sylvestre Duf our
described the fol l owi ng si t uat i on in his book Traitez nouveau
et curieus du cafe, du the, et du chocolat (1671): "Coffee sobers
you up i nst ant aneousl y, or in any event it sobers up those who
are not f ul l y intoxicated. One of my f r i ends who had had too
much wine sat down at the gambling table one evening a f t er
di nner . He was losing considerable sums, because of having
drunk too much wine, he was conf usi ng heart s with di amonds.
I took him aside and had him dri nk a cup of cof f ee, whereupon
he ret urned to the game with a completely sober head and
clear eye." (Schivelbusch, 1992).
We j us t learned t hat alcohol apparently raises adenosine lev-
186 / Buzz
els in the br ai n, whi l e caf f ei ne blocks adenosinc receptors and
could t hus pl ausi bl y reverse this ef f ect . Does this mean t hat
neuroscienec has v er i f i ed Duf our ' s t hr ee hundred-year-ol d ob-
servat i ons?
Not exact l y.
If adenosi ne was t he onl y thing t hat alcohol altered in the
br ai n, t hen c a f f ei ne would, i ndeed, neat l y count eract t hat ac-
tion and could be expected to reverse alcohol i nt oxi cat i on. But,
as we know, alcohol acts on many more brain systems t han j us t
adenosi ne. While it is af f ect i ng adenosine, alcohol is also mak-
ing GABA recept ors more sensi t i v e, and it's i nhi bi t i ng gluta-
mat e recept ors, rai si ng dopami nc levels, and exert i ng a wide
range of other compl i cat ed ef f ect s. Meanwhi l e, caf f ei ne can
only ant agoni / e adenosi ne recept ors. In a sense, caf f ei ne is
f i ght i ng with a single sword, while alcohol comes armed with
a do/ en weapons all fl ai l i ng at once.
It is est i mat ed t hat in general only 10 to 20 percent of al-
cohol' s i nt oxi cat i ng ef f ect can be at t r i but ed to increased aden-
osi nc levels ( Dunwi ddi c 1995). That means t hat even if you
dr ank enough caf f ei ne to plug every last adenosi ne receptor in
your brai n, you would not be staving off more than one- f i f t h
of alcohol-induced i nebri at i on. This is why one is well advised
to heed the popul ar wisdom t hat caf f ei ne will not of f set the
ef f ec t s of alcohol. But , as with most si t uat i ons involving these
two drugs, every rule has an except i on. When the amount of
alcohol circulating in the brain is low and the amount of caf-
f ei ne is high, the ant agoni sm of alcohol by caf f ei ne can be
si gni fi cant . In one st udy, 200 to 400 milligrams of caf f ei ne re-
versed poor per f or mance on some measures of driving abi l i t y
in subj ect s with blood alcohol levels of .04 percent to .06 per-
cent ( Moskowi t z and Bums 1981). Caf f ei ne has been shown
to rev erse alcohol-induced decrement s in flying-related ment al
and motor meas ur es and per f or mance on automobile Simula-
Better Living Through Chemistry I 187
t ors. Some st udi es of si mpl e reaction time have also shown that
c a f f e i ne erases the negative impact of alcohol (Fudin and Ni-
castro 1988).
Again, the critical caveat to all these studies is t hat the caf -
fei ne doses were al ways large rel at i ve to the alcohol doses. In
f act , the blood alcohol levels in subj ect s experiencing a reversal
of alcohol-induced per f or mance decrement s were all below the
.1 percent level t hat t ypi cal l y defi nes i nt oxi cat i on. No study-
has f ound t hat caf f ei ne rev erses the ef f ec t s of alcohol levels at
.1 percent or above (Fudin and Ni cast ro 1988). A few st udi es
have even f ound t hat when moderate to high levels of alcohol
are involved, caf f ei ne act ual l y worsens per f or mance on a v ar i et y
of reaction t i me and vi gi l ance t est s (Osborne and Rogers 1983).
Likewise, st udi es of cognitive per f or mance have shown that on
some t ypes of t est s, c a f f e i n e increased the del et eri ous ef f ec t s
of alcohol (Dews 1984).
The general sci ent i f i c consensus, t her ef or e, is t hat alcohol
and caf f ei ne i nt er act in complex ways t hat involve both antag-
onism and synergi sm, dependi ng on the dose of both drugs.
Ca f f ei ne most clearly of f s e t s t he di sabl i ng ef f ect s of alcohol
when the l ev el s of c a f f e i n e are high (above 200 milligrams) and
the l evel s of alcohol r el at i v el y low (below .1% blood alcohol
l ev el ). But even under t hese conditions, the reversal of alcohol' s
ef f ect s by caf f ei ne is incomplete. Although some of alcohol's
ef f ect s are count er act ed by caf f ei ne, others remain unt ouched.
A number of researchers have pointed out the potential clanger
of this s i t uat i on. Per haps t he br ai n' s sleep-regulating cent er
an area rich in adenosi ne recept orsi s one place where caf-
f ei ne best ant agoni z es t he ef f ec t s of alcohol. Ca f f ei ne could,
t her ef or e, make an i nt oxi cat ed person feel more alert even
though other part s of the brain arc considerably impaired. Driv-
ing a car or oper at i ng dangerous machinery under these con-
di t i ons would obviously be both irresponsible and ha/ ar dous.
S 88 / Buzz
The bottom l i ne is t hat Duf our ' s centuries-old observ at i on
i s bot h ri ght and wrong. He not ed, correct l y, t hat c a f f ei ne i s
most u s e f u l for t hose "not f u l l y i nt oxi cat ed. " But hi s compel-
l i ng descri pt i on of hi s dr unk f r i end' s mi r acul ous recovery t o a
"completely sober head and clear eye" a f t e r a si ngl e cup of
coffee was cl earl y a case of wi s hf ul t hi nki ng.
Engineering
We' ve now seen t ha t al cohol and caf f ei ne do not act in isola-
tion f r om each other at a mol ecul ar and neur onal level. By
af f ect i ng t he same key brai n neur ot r ans mi t t er , they are, i n fact ,
closely rel at ed. This synergy is refl ect ed at the behavioral level
as well.
People of t en use alcohol and caf f ei ne as compl ement ar y
tools for mood engi neer i ng. To par aphr ase a f amous adv ert i si ng
slogan for DuPont , t hey are the chemicals most of t en used to
achi ev e "better living. " The quot e by David Lct t crman at the
begi nni ng of this chapt er i l l ust r at es t he point. Lct t cr man
openl y and self-consciously consumes a lot of cof f ee, in part ,
he says, to induce the sl i ght l y manic comedic state for which
he is f a mous ( Zehme 1994). Among other professi onal s who
use caf f ei neand al cohol t o pri me their creative pumps, per-
haps the most wi del y known arc wr i t er s , many of whom have
provided eloquent t est i mony on t hi s pract i ce. In A Moveable
Feast, for i nst ance, Ernest Hemi ngway r ecal l s a t ypi cal day of
wr i t i ng in 1920s Pari s:
It was a pl easant caf e, warm and dean and f r i endl y, and I hung
up my old wat erproof on the coat rack to dry and put my worn
and weat hered fel t hat on the r ack above the bench and ordered
a cafe au l ai t . Th e wai t er brought it and I took out a notebook
from the pocket of the coat and a penci l and st ar t ed to wri t e. I
Belter Living 'Through Chemistry I ! 89
was wr i t i ng about up in Mi chi gan and since it was a wi l d, cold,
blowing day i t was t h a t sort of day i n t he st ory . . . in t he story
the boys were d r i n k i n g and t hi s made me t hi r s t y and I ordered
a rum St. James. This tasted wonder f ul on the cold day and 1
kept on wr i t i ng, f e e l i ng very well and f eel i ng the good Mar t i -
ni que rui n warm me all t hr ough my body and my spi r i t .
Hemingway, despite his consumpt i on of the rum St. James,
was proud of hi s abi l i t y t o separat e hi s l egendary dr i nki ng from
his writing (Dardis 1989). He generally drank only coffee while
he wrote and waited unt i l his notebook was closed for the day
bef or e i ndul gi ng i n al cohol .
John Steinbeck, too, usual l y confined hi msel f to caf f ei ne
when wri t i ng and relaxed wi t h alcohol. In the j our nal he kept
while wri t i ng The Grapes of Wrath, he made f r equent note of
the del et eri ous ef f ect alcohol had on his work (Demott 1989).
"Last ni ght up to Rays' and drank a great deal of champagne,"
he not ed bef or e t ur ni ng hi s at t ent i on t o hi s unf i ni shed man-
uscri pt on June 13, 1938. "I pulled my punches pr et t y well,
but I am not in the dead sober st at e I could wish." In the next
day' s ent r y he says: "Yesterday was a bust . I could have forced
the work out but I' d lost the flow of the book and it would
have been a weak spot."
F. Scott Fi t z geral d wrot e his most acclaimed books while
downing l ar ge hel pi ngs of caf f ei neusual l y in the form of cola
sodas and cof f ee. In his l at er years, he t ur ned to alcohol in in-
creasi ngl y desperate at t empt s to regain the muse. He began his
clays dri nki ng pots of cof f ee, and would t hen switch to bottles of
gi n i n t he a f t er noon. The r esul t i ng pharmacol ogi cal gyr at i ons
didn' t help: none of his l at er works is regarded as equal to The
Great Gatsby and his ot her early novels and short st ori es.
Hemingway, Steinbeck, and Fi t z geral d, of course, were al-
coholics, and t hus drank i n far greater quant i t i es and with
190 / Buzz
much great er i nt ensi t y t han most people. But their musings on
the importance of alcohol and caf f ei ne in their lives have res-
onance for many si mi l arl y inclined people. The urge to t i nker
wi t h one' s mood and energy level by using these yinyang drugs
is common. Common also is the experience of finding that
caf f ei ne simply exacerbates anxi et y, or t hat alcohol can inter-
f er e with emot i onal i nt i macy as well as fost er it.
To the ext ent t hat knowledge can i nf l uence behavior, the
i nf or mat i on presented in this book about how alcohol and caf-
f ei ne work might help people use these substances more ef f ec-
t i v el y and i nt el l i gent l y. But would Hemi ngway have dr unk
more moderat el y had he known how alcohol was af f ect i ng his
NMDA recept ors? Would Fitzgerald have tempered his cola
consumpt i on if he had known that the caf f ei ne was blocking
adenosi ne receptors in his brai n? It' s hard to imagine af f i r ma-
tive answers to these quest i ons. When it comes to drugseven
one as mild as caf f ei nel ogi c and reason can be i mpressi vel y
usel ess.
The Multi tudes Wi thi n
In "Song of Mysel f, " Walt Whitman wrote,
Do I cont radi ct mysel f ?
Very well. I cont radi ct myself.
I am large. I contain mul t i t udes.
Whi t man ant i ci pat ed by more than a hundr ed years a per-
spective on human nat ur e t hat i l l umi nat es t he of t en perplexing
rel at i onshi p people have with their drugs of choice.
The human brain is now known to be a layered and multi-
facet ed organ. It is subdi v i ded into discrete f unct i onal uni t s
t hat operate with a great deal of i ndependence. The brain, and
Better Living Through Chemistry I 191
the mind generat ed by the br ai n, have been likened to a "so-
ciety" of more or less autonomous parts (Minsky 1986). Viewed
from this perspective, it is not surpri si ng that when it comes
to drugs, humans are capable of pronounced contradictions.
The science of the mind is far less developed than the sci-
ence of the brain, and t hus st at ement s about how specific be-
haviors or cravings emerge from the workings of neurons are
necessarily qui t e tentative. But many neurosci ent i st s have spec-
ulated along the following lines. The neocortexthe most re-
cent addition to the human brain from an evol ut i onary point
of viewis the seat of language, music, abstraction, reason,
foresi ght , and refl ect i on. It is speculated that humans use their
neocortex to form their sense of who they aret hei r self-
awareness and their self-consciousness. The neocortex "un-
derstands" i nf or mat i on presented verbally, logically, and
sequentially: i nformat i on such as that presented in books about
the nat ure of alcohol and caf f ei ne, for instance. Other parts of
the brai n, however, do not work in this way. The limbic system,
for example, is believed to support emotions such as empathy,
anger, t er r i t or i al i t y, aggression, and mat ernal bonding. And
there are brain st r uct ur es that generate sexual desire, thirst,
hunger, pain, pleasure, and other pri mal sensations. All these
st r uct ur es are the neurological substrates of Whitman' s "mul-
titudes."
The conflict generated by the simultaneous activity of all
the members of the mind' s "society" is, of course, the f oun-
dation of much l i t erat ure and ar t . It is our capaci t y for i nt ernal
conflict and i rrat i onal i t y t hat defines us as human beings. Fic-
tional characters such as Data, the emotionless android of the
television show Star Trek: The Next Generation, are compelling
precisely because t hei r perfect logic and lack of emotion con-
trast so sharply with the very i mperfect logic of the humans
around t hem.
192 / Buzz
The poi nt i s t hat dr ugs such as alcohol and caf f ei ne af f ect
t he br ai n and mi nd at all levels. As we' ve seen repeat edl y, al-
cohol and c a f f ei ne go to work, ei t her di rect l y or i ndi r ect l y, on
t he nc ur ot r a ns mi t t c r s used i n t he ncoeort ex, t he l i mbi c system,
and t he dopa mi ne- f uc l ed reward cent ers, evoking very powerful
cr av i ngs and s e ns a t i ons t hat can collide wi t h, or completely
ov er whel m, more pr udent desi r es generat ed el sewhere i n t he
br ai n. An al cohol i c reaching yet agai n for a bot t l e despite the
knowledge t hat f u r t h e r dr i nki ng will be di s as t r ous is r espondi ng
not to r eason or logic, but to deeper voices ent i r el y. Likewise,
many cof f ee dr i nker s have experi enced t he t ug t o hav e anot her
cup even though t hey know f r om past exper i ence t hat yi el di ng
to the temptation is something t hey' re likely to regret.
None of t hi s means that increased knowledge is irrelevant to
one' s ef f or t s to rise alcohol or c a f f ei ne wisely. The fact remai ns
t hat most people are not addi ct ed to alcohol or caf f ei ne and
can control t hei r consumpt i on to one degree or anot her. They
are nei t her compl et el y capt i v e to t hei r cravings nor so in con-
trol t hat t hey don' t occasionally dri nk more alcohol or c a f f ei ne
t han they know is h e a l t h y or product i v e. This suggests t hat
i nf or mat i on about how al cohol and c a f f ei ne work will be us ef ul
to v aryi ng degrees for d i f f e r e n t i ndi v i dual s. A more complete
under st andi ng of t hes e subst ances may help people find ways
t o use them more ef f ect i v el y i n t hei r dai l y l i v es. Modern neu-
roscience suggests, however, t hat it would be a mistake to dis-
count t he mu l t i pl i c i t y of t he mi nd, t o forget t ha t one' s
consci ous self is not one' s ent i r e sel f , and to ignore the power
of t he nonr at i onal forces wi t hi n us.
This deep di chot omy between reason and i r r at i onal i t y can
be seen in t he wor l d' s t r emendous appet i t e for alcohol and
c a f f e i ne . Alcohol is t he l i berat or of the i r r at i onal . Ca f f ei ne is
t he s t i mul at or of t he r at i onal . It woul d appear t hat t he human
spirit craves both poles and t u r n s to t hese most f a mi l i a r of
drugs t o achi ev e t hose ends .
Having now wri tten an enti re book about alcohol and caf f ei ne,
and having mused about thei r uti l i ty for the enhancement of
dai l y living, 1 f eel that I can't ignore an obvious topic: the role
these two substances played in the creati on of this book and,
conversely, the ef f ect the book has had on my use of these two
substances.
Most of Buzz has been wri tten on caf f ei ne. As 1 write thi s
sentence, i t's 11:12 P.M. I would be asl eep were it not for the
caf f ei ne molecules coursi ng through my system. I would, in
f act, prefer to be asleep. But ci rcumstances have forced me to
pursue thi s book in the wee hours of my l i f e. Thus caf f ei ne,
194 I Buzz
t ypi cal l y admi ni s t er ed in a shot or two of ev eni ng espresso, has
been an i nv al uabl e t ool . As an exper i ment , I have t ri ed wri t i ng
without c a f f e i ne at t i mes l i ke t hi s , and t he resul t s are not
pr et t y. The wr i t i ng comes out j us t as t i r ed and flabby as I feel
physi cal l y.
In cont r ast , al cohol has pl ayed onl y an i ndi r ec t role in t hi s
book. My abi l i t y t o wr i t e i s t h e f i r s t t hi ng t o di ssol v e i n alco-
hol' s solvent, if I have as little as a hal f - gl ass of wi ne an hour
before wri t i ng some cri t i cal pr essur e i s lost. With alcohol in
my system, I cannot , as Hemi ngway once said, "close it like
the di aphr agm of a camera and i nt e ns i f y it so it coul d be con-
cent rat ed to the poi nt where the heat shone bright and t he
smoke began to rise. "
1 am not , however, a t eet ot al er . Of t e n, a f t e r wr i t i ng, I in-
dulge in a dram of my f av or i t e scotch or a smal l shot of good
bourbon. And al t hough I forgo wi ne on ni ght s when I write, I
gr eat l y enj oy t he gi ft of Bacchus when mat ed wi t h t he right
food. Alcohol, in other words, is a nor ma l and enj oyabl e part
of my l i f e and t hus probabl y deserves ment i on as playing some
kind of support i ng role in the book' s creat i on.
But t he book has a f f ec t ed my use of al cohol and c a f f ei ne
j us t as much as my use of alcohol and c a f f ei ne has a f f ec t ed
the book. In the past, I somet i mes reached for a cup of coffee
or a glass of wi ne as much f r om habit as f r om a conscious desire
to alter my consci ousness. Beverages cont ai ni ng alcohol and
caf f ei ne are so embedded in modern society t hat it is easy to
forget t hat t hey cont ai n r el at i v el y powerful dr ugs. Now, of
course, I can' t i gnore t hi s f act , and this has made me a more
conscious consumer.
In general , I dr i nk less alcohol now than I did pri or to be-
gi nni ng t hi s pr oj ect . I don' t aut omat i cal l y t ake a glass of wi ne
pr of f er ed at a part y, or assume t ha t if I meet f r i e n d s at a bar
I need to hav e a beer. In shor t , I t ry, with v ar yi ng degrees of
Postscript I 195
sueeess, to use alcohol del i berat el yt o enj oy it when the oc-
casion war r ant s it, and to avoid it when I want a sharper, clearer
state of mind.
As for caf f ei ne, I continue my long-standing experiments to
find an optimal dose. I love the flavors, the aromas, and the
rituals surroundi ng good coffee and espresso, but I' m also
aware that a caf f ei ne buz z is us ef ul for some tasks and not for
others. When patience and calm are requi red (such as when
caring for small chi l dren), I have found caf f ei ne to be of du-
bious ut i l i t y. When the job is clear-cut, or when a slightly
manic f r ame of mind is enjoyable, caf f ei ne can be j us t the
ticket. To remind myself of the ways caf f ei ne af f ect s my mood
and personality, I take occasional caffei ne holidays of at least
a week or two. Knowing how caf f ei ne works in the br ai n has
allowed me to tailor my i nt ake to mi ni mi z e unpl easant with-
drawal symptoms.
In short, two years of research and wri t i ng about alcohol and
caf f ei ne hasn' t convinced me to stop consuming either drug.
If I were to boil down the contents of t hi s book to a few in-
tensely-flavored drops of advice, I believe I'd end up with some-
thing very si mi l ar to some bits of wisdom carved more t han
2000 years ago into the stone face of the temple of Apollo at
Delphi. Two simple phrases were etched so deeply t hat to this
day you can still read them easily: "know t hysel f, " and "nothing
to excess." If the current scientific underst andi ng of alcohol
and caf f ei ne says anything, it is exactly t hat .
This page intentionally left blank
Amcri nc, VI. A., and 1'j. B. Rocsslcr. 1983. Wines: Their Sensory Evaluation, New
York: Fr eeman.
Amit, Z., E. A. Sut her l and, K. Gill, and S. O. Ogrcn. 1984. Zi ni el i dme: A review
of its e f f e c t s on alcohol consumpt i on. Neuroscience and Kiobehavioral Re-
views 8:^5-54.
Ar naud, VI. J. 1993. Met abol i sm of c a f f e i n e and other component s of coffee. In
Caffeine, Coffee, and Health, edi t ed by S. Gar at t i ni . New York: Raven.
Ashton, II. 1992. Brai n Funct i on and Psychotropic Drugs. New Yor k: Oxf or d U n i -
v er si t y Press.
At hanas i on, R., P. Shaver, and C. Tavri s. 1970. Sex: A Psychology 'iodcty report on
more t han 20,000 responses t o 101 quest i ons about sexual a t t i t udes and
pr act i ces. Psychology Today, February: 39-52.
Bar r , I I . VI., and A. P. St rei ssgut h. 1991. Caf f ei ne use dur i ng pregnancy and chi l d
198 / References and Suggested Reading
out c ome : A 7year pr ospect i v e st udy. Neurotoxicology and Teratology
13:441-448.
Barret t -Connor. L., j. C. Chang, and S. L. Fdcl s t cm. 1994. Coffee-associ at ed os-
t eoporosi s of f set by dai l y mi l k consumpt i on. Journal of the American Med-
ical Association 271:280-283.
Battig, K., and 11. \Vetzl. 1993. Psychopharmacological profile of caf f ei ne. In Caf-
feine, Coffee, and Health, edi t ed by S. Car at t i ni . New York: Rav en.
Benowi t z , N. L., S. M. Hal l , and C. Moclin. 1989. Per si st ent i ncrease in caf -
f ei ne c onc ent r a t i ons in people who s t op smoki ng. British Medical Journal
298:1075-1076.
Ber r y, 11., Beverage Mar ket i ng Cor por at i on, New York 1994. Personal communi-
cat i ons.
Bi r n b a n m, I. M. , E. S. Par ker , and ]. T. Har t l ey. 1978. Alcohol and h u ma n mem-
ory: Ret r i ev a l processes. Journal of Verbal Learning and Verbal Behavior
17: 325-335.
Bl i t z cr, R. IX 1994. Personal communi cat i ons .
Bl i t / er , R. D., C). Gil, and E. M. Landau. 1990. Long-term pot cnt i at i on in rat
hi ppocampus is i nhi bi t ed by low concent r at i ons of alcohol. Brain Research
537:203-208.
Bur ke, I: ; . R. 1992. Cycling Health and Physiology. Br at t l ebor o, Vt: Vi t esse Press.
Caf f ei ne as a downer: Rare r eact i on to c a f f e i n e . 1990. Harvard Medical School
Health Letter, Apr i l : ?.
Char ness, M. K. 1994. Personal communi cat i ons ,
Char ness, M. E., R. M. Sa f r a n , and C. Pericles. 1994. Et hanol i nhi bi t s ne ur a l ccll-
eell adhes i on, journal of Biological Chemistry 269:9304-9309.
Charness, M I
1
' , . , R. P. Si mon, and 1). A. Creenberg. 1989. Alcohol and the ner v ous
syst em. New England Journal of Medicine 321: 4424?4.
Chen, C.-C., and E. -K. Ych. 1989. Popul at i on di f f er ences in ALDI I l ev el s and
flushing response. In Molecular Mechanisms of Alcohol, edi t ed by C. Y. Sun.
New Yor k: l l nmana.
Chou, T. 1992. Wake up and smell the cof f ee: Ca f f e i n e , coffee, and t he medi cal
consequences. Western Journal of Medicine 157 : 5445^ 4.
Cicero, 'L. ]. 1994. Ef f ect s of pat er nal exposure to alcohol on of f s pr i ng develop-
ment . Alcohol Health and Research World 18:37-41.
Cloninger, C. R., M. Bohman, and S. Siguardsson. 1981. Inheri t ance of alcohol
abuse: Cross-fost eri ng a na l ys i s of adopt ed men. Archives of General Psychi-
atry 38:861-868.
Cohen, S., D. A. Tyrrell, M. A. Russel l , J . ) . Jams, and A. P. S mi t h . 1993. Smoking,
alcohol consumpt i on, and suscept i bi l i t y to t he common cold. American
Journal of Public Health 83:1277-1283.
Coles, C, 1994. Cri t i cal per i ods for pr enat al alcohol exposure. Alcohol Health and
Research World 18:22-29.
Costill, I). L., C. Dal asky, and W. l i nk. 1978. Ef f ect s of caf f ei ne ingcstion on
met abol i sm and exercise per f or mance. Medicine and Science in Sports and
Exercise 10:155-158.
References and Suggested Reading I 199
Crabbc, J. C., J. K. Bcl knap, and K. J. Buck. 1994. Genetic a ni ma l model s of al cohol
and dr ug abuse. Science 264:1715-1723.
Daly, J. W. 1993. Mechani sm of act i on of c a f f e i n e . In Caffeine, Coffee, and Health.
edited by S. Gar at t i ni . New York: Raven.
Daly, J. W., D. Shi, V. Wong, and O. Ni kodi j cv i e. 1994. Chroni c ef f ect s of ct hanol
on central adenosine f unct i on of mice. Brain Research 650: 153156.
Dardis, T. 1989. The Thirsty Muse. New York: Ti cknor & Fi el ds.
Demot t , R. 1989. Working Days: The journals of "The Crapes of Wrath." New-
York: Penguin.
Depar t ment of Heal t h and Human Ser v i ces, Nat i onal I n s t i t u t e on Alcohol Abuse
and Alcoholism. 1993. Alcohol and Health: Eighth Special Report to the
V. S, Congress.
Dews, P. B. 1984. Caffeine: Perspectives from Recent Research. New York: Spri uger-
Verlag.
Di Chiara, G., and A. I mper at o. 1988. Drugs abused by humans pr ef er ent i al l y
increase synapt i c dopamme concent r at i ons m the mesohmbi c syst em of
f r e e l y moving r at s . Proceedings of the National Academy of Sciences 85: 5274-
5278.
Dolnick, E. 1990. Le par adox f r a nga i s . Health, May-June:41-47.
Dowling, J. E. 1992. Neurons and Networks. Cambri dge, Mass. : Harv ard U ni v er s i t y
Press.
Dumas, VI. 1982. Syst emat i c det er mi nat i on of c a f f ei ne plasma concent r at i ons at
birth in prc-t erm and f ul l - t er m i nf a nt s . Developmental Pharmacology and
Therapeutics 4:182-186.
Dunwi ddi c, T. 1995. Personal communi cat i ons.
Fskena/ i , B. 1993. Caf f ei ne dur i ng pregnancy: Gr ounds for concern? Journal of the
American Medical Association 270:2973-2974.
Ev ans, S. M. , and R. R. Gr i f f i t hs . 1992. Ca f f e i ne t ol erance and choi ce in humans .
Psychopharmacohgy 108:51-59.
Fastbom, J. , and B. B. Fredholm. 1990. Effects of long-term thcophvllinc t reat ment
on adenosine A, recept ors in rat bram: Aut oradi ographi c ev i dence for in-
creased recept or number and al t ered coupl i ng to G- pr ot ei ns. Brain Research
507:195-199.
Fredholm, B. B. 1984. Ef f e c t s of met hyl xant hi ucs on skel et al mus cl e and on iex-
pi rat i on. In The MethyUanthine Beverages and Foods: Chemistry, Consump-
tion, and Health Effects, edi t ed by G. A. Spi l l er . New York: l i ss.
Fre/.za M., C. Di Padov a, G. Po/.zato, M. Tcrpin, F. Bar aona, and C. S. Li cber .
1990. High blood alcohol l ev el s in women: The role of decreased ga s t r i c
alcohol dehydrogcnasc act i v i t y and f i r st - pass met abol i sm. New England Jour-
nal of Medicine 322:95-99.
Fri edman, G. D., and A. L. Kl at sky. 1993. Is alcohol good for your heal t h? New
England Journal of MeJicme'329:1882-1883.
Fri edman, H. ]., J. A. Car pent er , D. Lester, and C. L. Randal l . 1980. Ef f ec t of
al pha-met hyl -p-t yrosi ne on dose-dependent mouse s t r ai n di f f e r e nc e s in lo-
comofor act i v i t y a f t e r alcohol. Journal of Studies on Alcohol 41: 1-7.
200 / References and Suggested Reading
Fryc, G. D., and G. R. Brccsc. 1981. An ev al uat i on of the locomotor s t i mu l a t i n g
act i on of alcohol in r at s and mice. Psychophannacology 75:372-379.
Eudi n, R. , and R. Nicastro. 1988. Can c a f f ei ne ant agoni z e al cohol - i nduced per f or -
mance decrement s in h u ma n s ? Perceptual and Motor Skills 67:375-391
Gallctly, O. C., M. Fcnncl l y, and J. C. Whi t wam. 1989. Docs c a f f ei ne wi t hdr awal
cont nhut e t o post auaest bchc mor bi di t y
7
Lancet 1 : 1 335.
Garci a, R. 1993. The cardi ov ascul ar ef f ect s of c a f f e i n e . In Caffeine, Coffee, and
Health, edited by S. Ga r a t t mi . New York: Rav en.
Gi anoul aki s, C., P. Angel ogi anni , M. Meany, J. Th a v u n d a y i l , and V. Tawar. 1990.
Fndorphi ns in i ndi v i dua l s wi t h hi gh and low ri sk for development of alco-
hol i sm. In Opioids, Bulimia, and Alcohol Abuse and Alcoholism, edi t ed by
L. D. Reid. New York: Spri nger-Vcrl ag.
Gilbert, R. M. 1984. Ga f f e me cons umpt i on. I n ihe Methylxanlhine Beverages and
Voods: Chemistry, Consumption, and Health Infects, edi t ed by G. A, Spi l l er.
New York: Liss.
Gi l bert , R. M. 1992. Caffeine, the Most Popular Stimulant. New York: Chel sea
Mouse.
Gi t t el man, R. 1983. Expe r i me nt a l and cl i ni cal st udi es of s t i mul ant use in hyper-
act i v e chi l dr en and chi l dr en with ot her behav i or al di s or der s . In Stimulants:
Neurochemical, Behavioral and Clinical Perspectives, edi t ed by 1. Creese.
New York: Rav en.
Glass, R. M. 1994. Ca f f e i n e dependence: What arc t he i mpl i cat i ons ? journal of
the American Medical Association 272: 10651066.
Gol dst ei n, A. 1994. Addiction, from Biology to Drug Policy. New York: Freeman.
Goodwi n, D. W. 1985. Al cohol i sm and genet i cs: The s i ns of t he f a t h e r s . Archives
of General Psychiatry 42:171-174.
Gordis, E., M. C. Duf our , K. R. Warren, R. J. Jackson, R. L. Floyd, and 13. W.
Hungcrford. 1995. Should physi ci ans counsel pat i ent s to dr i nk alcohol? Journal
of the American Medical Association 273: 1415.
Gordis, L. 1990. Cons umpt i on of met hyl xant hi ne- cont ai mng beverages and risk of
pancr eat i c cancer. Cancer Letters 52:1-12.
Gordon, A.. M. K. Sapru, S. W. Kr aus s , and I. Di amond. 1993. Nuclcosidc t r ans-
port and et hanol - mduccd hct cr ol ogous cl escnsi hz at i on. I n Alcohol, Cell
Membranes, and Signal Transduction. in Brain, edi t ed by C. Ailing ct al .
New York: Pl enum.
Gr aham, I I . N. 1984. 'Pea: The pl a nt and its ma nuf a c t ur e : Chemi st r y and con-
sumpt i on of t he beverage. In The Methyhanthine Beverages and Foods:
Chemistry, Consumption, and Health Kffects, edi t ed by C. A. Spi l l er . New
York: Liss.
Grobbcc, D. K. , E. B. Ri mm, K. Gi ov ami ucci , G. Colditz, M. St ampf cr , and W.
Wi l l ct t . 1990. Cof f ee, c a f f ei ne, and car di ov as cul ar disease i n men. New
England Journal of Medicine 323: 1026- 1032.
Gr ounds for br eaki ng the coffee habi t ? 1990. 'Tufts University Diet (5- Nutrition
Letter, February: ?.
Ma r b u r g , E., D. R. Dav i s, and R. Gapl an. 1982. Par ent and o f f s pr i n g alcohol use:
References and Suggested Reading I 201
I mi t a t i v e and av ersi v c t r ansmi ssi on. Journal of Studies on Alcohol 43: 497516.
Harper C. G., and J. J. Kr i l l . 1990. Neuropat hol ogy of alcoholism. Alcohol and
Alcoholism 25:207-216.
Hignci, S., S. Mat sushi t a, I I . Ima/ eki , 'I'. Ki noshi t a, S. Takagi, and H. Kono. 1994.
Al dehyde dchydrogcnasc genot ypes in Japanese al cohol i cs. Lancet 343:741-
742.
Holdeii, C. 1994. A caut i onar y genetic tale: The soberi ng st ory of D
7
. Science
264:1696.
Hol l i st er, L. K. 1975. The mys t i que of social drugs and sex. In Sexual Behavior,
Pharmacology, and Biochemistry, edi t ed by M. Sandier and G. L. Gessa. New
York: Rav en.
Hooper, J . , and D. Teresi. 1987. 'I'he Three-Pound Universe. New York: Del l .
I l or gan, J. 1993. Eugeni cs r ev i si t ed. Scientific American, J u n e : 12> ~ 131.
Hughes, J. R. 1992. Cl i ni cal i mpor t ance of c a f f e i ne wi t hdr awal . New England jour-
nal of Medicine 327: 1160-1161.
I nf ant e- Ri v ar cl G., A. Fe r n a n d e z , R. Gaut hi er , M. Dav i d, and C. E. Ri v av cl . 1993.
Fetal loss associated wi t h caf f ei ne i nt ake before and dur i ng pregnancy, jour-
nal of the American Medical Association 270:2940-2943.
I v er sen, S. D., and L,. L. Iv crscn. 1981. Behavioural Pharmacology. New York: Ox-
f or d Uni v er s i t y Press.
Jacobson, J. I, . , and S. W. Jacobson. 1994. Pr enat al alcohol exposur e and ncur o-
behav i or al dev el opment . Alcohol Health and Research World 18:30-36.
Jacobson S. W., G. G. Fein, J. L. Jacobson, P. M. Schwart / , , and J. K. Dowler.
1984. Neonat al correl at es of prenat al exposure to smoki ng, c a f f e i ne , and
alcohol. Infant Behavioral Development 7:253-265.
James, J. 1991. Caffeine and Health. New York: Academic Press.
Jensen, G. B., and B. Pakkenbcr g. 1993. Do al cohol i cs dri nk t hei r neurons away?
Lancet 342: 1201-1204.
Johnson, G. 1991. In the Palaces of Memory. New York: Knopf .
Jones, B. M. 1973. .Memory i mpai r ment on the ascendi ng and descendi ng hmbs
of t he blood alcohol cur v e, journal of Abnormal Psychology 82: 24-32.
Jossel yn, S. A., and R. J. Bcni nger. 1991. Behav i oral e f f e c t s of i n t r a s t r i a i a l c a f f e i n e
medi at ed by adenosmer gi e modul at i on of dopamme. Pharmacology, Bio-
chemistry and Behavior 39:97-103.
Ka pl a n, C. 1995. Personal c ommuni c a t i ons .
Kapl an, G., D. J. Gr eenbl at t , M. A. Kent , and M. M Cot reau-Bi bbo. 1993. Ca f f e i ne
t r eat ment and wi t hdr a wa l in mice: Rel at i onshi ps bet ween dosage, concen-
t r at i ons , locomotor a c t i v i t y and A, adenosi nc receptor bi ndi ng, journal of
Pharmacology and Experimental 'I'herapeulics 266:1563-1572.
Kami , A., D. Tannc, B. S. Rnbcnst cm, J. J. M. Askenasv , and I ) . Sagi. 1994. De-
pendence on RFA'l sleep of overnight i mpr ov ement of a percept ual s ki l l .
Science 265:679-682.
Kendl er K., A. C. Heat h, M. C. Neale, R. C. Kessl cr , and I, J Eav es, 1992. A
popul at i on- based t wi n s t udy of alcoholism in women, journal of the Amer-
ican Medical Association 268:1877-1882.
202 / References and Suggested Reading
Ki l i l r nan, B. A. 1977. Caffeine and Chromosomes. Oxf or d: Elscvicr.
Koob, G. R. , and I
1
' . li. Bl oom. 1988. Cel l ul ar and mol ecul ar mechani sms of dr ug
dependence. Science 242:715-719.
Kur ppa, K, P. C. l i ol mber g, V,. Kuosma, and I , . Saxcn. 1983. Cof f ee consumpt i on
dur i ng pr egnancy and sel ect ed congeni t al ma l f or ma t i on: A nat i onwi de case-
cont r ol st udy. American Journal of Public Health 73: 1397-1399.
Lang, A. K. , D. J. Coeckncr, V. J. Aclcsso, and G. A. Mar l at t . 1975. Ef f ect s of
al cohol on aggression in mal e soci al dr i nke r s . Journal of Abnormal Psychology
84:418-425."
Lang, A. R., J. Scarlcs, R. Laucr nr an, and V. Adcsso. 1980. Expect ancy, alcohol,
and sex gui l t as d e t e r mi n a n t s of i nt er es t rn and reaction t o sexual st i mul i .
Journal of Abnormal Psychology 89:644-653.
Laska L., A. Sunshi ne, F. Muel l er, W. B. Elvers, C. Sicgcl, and A. Rubi n. 1984.
Ca f f e i n e as an anal gesi c a d j u v a n t . Journal of the American Medical Associ-
ation 251: 1711- 1718.
La wr i n, M. ()., C. A. Na r a nj o, and E. M. .Sellers. 1986. I dent i f i c a t i on of new drugs
for modul a t i ng al cohol cons umpt i on. Psycho-pharmacology Bulletin 22:
1020-1025.
Lehman, 15. A. 1989. A t empest in a cof f ee pot . Boston Globe, November 27: 25.
Label, R. I , . , iVl. Rosenbaum, and J. I l i r s c h. 1995. Changes i n energy expendi t ur e
r es ul t i ng f r om a l t er ed body wei ght . New England Journal of Medicine 332:
621-628.
Lcwin, L. 1931 11964]. Phantastica: Narcotics and Stimulating Drugs. New York:
Out t o n .
Li v crmorc, 15. 1991. Caf f ei ne boosts eat i ng di sorders. Health, Junc: 16.
Lovi nger, D. VI., and R. VV. Peoples. 1993. Act i ons of alcohols and other sedative/
h y pn o t i c compounds on cat i on channel s associated with gl ut anr at e and
5-I1T3 recept ors. In Alcohol, Cell Membranes, and Signal Transduction
in Brain, edi t ed by C. Ai l i ng et al. New York: Pl enum.
Makowsky, C. R., and P. C. Whi t ehcad. 1991. Adv ert i si ng and alcohol sales: A
legal i mpact st udy. Journal of Studies on Alcohol 52: 555-567.
Mann, C. 1994. Behav i or al genet i cs in t r a ns i t i on. Science 264:1686-1689.
Ma s t e r s , W. M. , and V. E. Johnson. 1986. Masters and Johnson on Sex and Human
Laving. Bost on: Li t t l e, Brown.
McCl car n, C. F.., and D. A. Rodgcrs. 1959. Di f f er enc es in alcohol pr ef er ence
among i nbr ed s t r a i ns of mice. Journal of Studies on Alcohol 20:691-695.
McCoy, E., and J. F. Walker. 1991. Coffee and Tea. New York: Theron Rai nes.
Meadows, C. G., S. E. Bl ank, and D. D. Duncan. 1989. I nf l uence of alcohol con-
s umpt i on on n a t u r a l ki l l er cell act i v i t y in mice. Alcoholism: Clinical and
Kxperimental Research 13:476-479.
Mi l l er , M. 1994. Call for a dai l y dose of wi ne f er ment s cr i t i cs. Wall Street Journal,
J u n e 17: Bl .
Mi l l s J. L., L. 15. Hol mes, J. II. Aarons, et al . 1993. Moderat e c a f f ei ne use and t he
r i s k of spont aneous abort i on and i n f r a u t c r i n c growth r et ar dat i on. Journal of
the American Medical Association 269:593-597.
Mi nsk} -, M. 1986. The Society of Mind. New York: Simon and Schuster.
References and Suggested Reading I 203
Moskowi t / , I I . , and M. Burns. 1981, The e f f e c t s of alcohol and c a f f ei ne, alone and
in combi nat i on, on skills per f or mance. In Alcohol, Drugs and Traffic Safety,
edi t ed by L. Goldberg. Stockholm: Al mqv i st & Wiksel.
Murphy, J. M., W. J. McBri de, L. Lumcng, and T. -K. Li. 1987. Cont ent s of
monoarnmes in forcbram regi ons of al cohol -preferri ng and non- pr ef er r i ng
hues of r at s . Pharmacology, Biochemistry and Behavior 26:389392.
Na r a nj o, C. A., K. E. Kadlec, P. Sanhuc/ a, D. Woodley-Rcmus, and E. M. Sel l ers.
1990. El uoxct i ne di f f er ent i a l l y alters alcohol i nt ake and ot her consmiima-
tory behaviors in problem dr i nker s. Clinical Pharmacology and Therapeutics
47:490-498.
Nash, [I . 1962. Alcohol and Caffeine: A Study of Their Psychological Effects. Spring-
f i el d, I I I . : Thomas.
Nat hanson, ]. 1984. Caf f ei ne and related mct hyl xant hi nes: Possible nat ur al l y oc-
cunng pest i ci des. Science 226: 184187.
Nat i onal Cof f ee Associ at i on. 1991. Cof f ee- dr i nki ng st udy.
Nchlig, A., J. L. Daval, and C. Debry. 1992. Ca f f ei ne and the cent ral nervous
syst em: Mechani sms of action, biochemical, metabolic and psychost i mul ant
ef f ect s . Brai n Research Reviews 17:139-170.
N1H consensus dev el opment panel on opt i mal cal ci um i nt ake. 1994. journal of the
American Medical Association 272:1942-1948.
Noble, E. P., K. Bl um, R. Ri t chi e, A. Montgomery, and P. J. Sher i dan. 1991. Allehc
associ at i on of t he D
2
dopammc receptor gene wi t h receptor bi ndi ng char-
act eri st i cs in al cohol i sm. Arc/jives of General Psychiatry 48:648-654.
Osborne, D. ]., and Y. Rogers. 1983. I nt er act i ons of alcohol and caf f ei ne on human
reaction t i me. Aviation, Space and Environmental Medicine 54: 528534.
Palca, ]. 1989. Sleep r esear cher s awake to possibilities. Science 245: 351.
Parsons, W. D., and A. I I . Nemi s. 1978. Ef f ect of smoking on c a f f e i ne clearance.
Clinical Pharmacology and Therapeutics 24:4045.
Phi l l i s, John W. 1989. Caf f ei ne and pr emenst r ual syndrome. American Journal of
Public I l eal t h 79:1680.
Pickens, R. W., D. S. Svikis, M. McGuc, D. T. Lykken, L. L. Hcst en, and P. J.
Cl ayt on. 1991. Het erogenei t y in the i nher i t ance of alcoholism. Archives of
General Psychiatry 48:19-28.'
Proctor, W. R, , and T. V. Dunwi ddi e. 1984. Behavioral sensi t i v i t y to puri nergi c
drugs parallels cthanol s ens i t i v i t y in selectively bred mice. Science 224:519-
521.
Rai nni e, D. G., M. Grume, R. W. McCarley, and R. W. Greene. 1994. Adenosme
i nhi bi t i on of rnesopont i ne eholinergic neurons: Impl i cat i ons for KEG
arousal . Science 263:689-692.
Regestem, Q. 1995. Personal communi cat i ons.
Reut ers News Service. 1994. Germany' s Gcrasch banned for two years. J a nua r y 26.
Ri char ds, B. 1995. Cafe au rev oi r? Some say cof f ee has become too cool. Wall
Street Journal, J a nua r y 31: A1.
Ri st o R., R. T. Gent r y, R. I I cr nandcx- Muno/ , E. Baraona, and C. S. Lieber. 1990.
Aspi ri n increases blood alcohol concent r at i ons in humans af t er ingcstion of
ct hanol . Journal of the American Medical Association 264:2406-2408.
204 / References and Suggested Reading
Robson, R. A. 1992. The ef f ect s of qui nol i nc s on x a n t h i u c pbar macoki net i cs . Amer-
ican Journal of Medicine 92: 22.
Rossignol, A. M. , II. Ronnl ancl cr , L. Song, and J. W. Phi l l i s. 1991. Do women wi t h
pr e me ns t r ua l s yni pt oui s s el f - medi c a t e wi t h c a f f ei ne? Epidemiology 2:403-
408.
Sal v at ore, C. A. 199?. Mol ecul ar cl oni ng and char act er i z at i on of t he h u ma n A,
adcnosme recept or. Proceedings of the National Academy of Sciences. 90:
10365-10369.
Sanders, R. C., and T. R. Mu r r a y . 1988. Chr oni c f heophyl l mc exposure i ncreases
agonist and ant agoni st bi ndi ng to A, adcnos i ne receptors in rat br ai n. Neu-
ropharmacology 27:757-760.
Saxcna, Q., R. Saxcna, and W. Adl cr. 1981. Regul at i on of n a t u r a l ki l l er act i v i t y in
vivo: High nat ur al ki l l er a c t i v i t y in al cohol -dri nki ng mi ce. Indian journal of
Experimental Biology 19:1001-1006.
Schi v cl busch, W. 1992. Tastes of Paradise: A Social History of Spices, Stimulants,
and Intoxicants. New York: Pa nt he on.
Schucki t , M. A. 1985. The cl i ni cal i mpl i c a t i o n s of pr i ma r y diagnostic groups among
alcoholics. Archives of General Psychiatry 42:1043-1049.
Scale T. W., K. A. Abl a, M. T. Sha i mm, J. M. Car ney, and J. W. Dal y. 1988. 3,7
di mc t l i yl - l - pr opa r gyl xa nt l i i nc : A pot ent and selective in vivo ant agoni st of
acl cuosi nc anal ogs. Life Sciences 43:1671-1684.
Sci t z , I I . , G. tigcrcr, and U. Si ma nows ki . 1990. Hi gh blood al cohol l evel s i n women.
New England journal of Medicine 323: )8.
Shi, D., (). Ni kodi j c v i c , K. A. Jacobson, and ). W. Dal y. 1993. Chr oni c c a f f ei ne-
alters the densi t y of adcnosi ne, adrencrgi c, chol i ucrgi e, GABA, and seroto-
nin receptors and cal ci um channel s in mouse br ai n. Cellular and Molecular
Neurobiology 13:247-261.
Siegcl, R. K. 1989. Intoxication: Life in Pursuit of Artificial Paradise. New York:
Dut t on.
Si l v cr ni an, K., S. M. Evans, K. C. St r a i n, and R. R. Gr i f f i t hs . 1992. Wi t hdr awal
syndr ome a f t e r t he doubl e- bl i nd cessat i on of c a f f e i n e cons umpt i on. New
England journal of Medicine 327: 1109-11 14.
Smar t , R. G. 1988. Does alcohol a dv e r t i s i ng a f f e c t ov er al l cons umpt i on? A review
of empi ri cal s t udi es . Journal of Studies on Alcohol 49:314-323.
Sncl , J. 1993. Sleep and wa kc f ul ues s . In Caffeine, Coffee, and Health, edi t ed by S.
Gar at t i ni . New York: Rav en.
St r adhng, J. R. 1993. Recr eat i onal drugs and sleep. British Medical journal
306: 573.
St r a i n, K C., G. K. Mi mi f o r d , K. Si l v er man, and R. R. Gr i f f i t h s . 1994. Ca f f ei ne
dependence s yndr ome, journal of ihe American Medical Association
272: 1043-1048.
Takes hi t a, T., K. Mor i mot o, X.-Q. Mao, T. Ha s h i mo t o , and J. Fur yna ma . .1993.
Phcnot ypi c di f f er enc es i n l ow K
m
al dehyde dchydr ogcnase i n J apanes e work-
ers. Lancet 341:837-838.
Tanaka, Y. 1990. Ef f e c t of x a n t h mc der i v a t i v es on hi ppocampal l ong-t erm pot cu-
t i a t i o n . Brain Research 522:63-68.
References and Suggested Reading I 205
Tice, P. VI. 1992. Altered States: Alcohol and Other Drugs in America. Rochest er,
N.Y.: Strong Museum.
Treistman, S. NT. , M. M. Moymhan, and D. E. Wolf. 1987. I nf l uence of al cohol s,
t emper at ur e, and region on the mobility of hpi ds i n neuronal membrane.
Riochwnca et Riophysica Acta 898:109-120.
Uhl , G., K. Blum, E. Noble, and S. Smi t h. 1993. Subst ance abuse v ul ner abi l i t y and
0, receptor genes. 'Trends in Neurosciences 16:83-88.
Vi ani , R. 1993. The composition of c of f ee. In Caffeine, Coffee, and Health, edited
by S. Ga r a t t i ni . New York: Raven.
Weight, F. E. 1994. Personal communi cat i ons.
Weight, F. F. 1992. Cellular and molecular physi ol ogy of alcohol act i ons in the
ner v ous syst em. International Review of Neurobiology 33:289-348.
Weight, F. F., R. W. Peoples, J. M. Wri ght , C. Li, I, G. Aguaya, D. M Lovingcr,
and G. Whi t e. 1993. Ncur ot r ansmi t t er - gat cd ion channel s as mol ecul ar si t es
of alcohol act i on. In Alcohol, Cell Membranes, and Signal Transduction in
Brain, edited by C. Ailing et al. New York: Pl enum.
Wei r, J. , T. D. Noakcs, K. Myburgh, and B. Adams. 1987. A high carbohydrat e
diet negates the metabolic effects of caffeine during exercise. Medicine and
Science in Sports and Exercise 19: 100105.
Wi l ford, J. N. 1992. Jar in I r ani an r ui ns bet rays beer dr i nker s of 3500 B.C. New York
Times, February 13: A16.
Wilson, G. T., and D. N4. Lawson. 1976. Expect anci es, alcohol, and s exual ar ousal
in mal e social dr i nker s . Journal of Abnormal Psychology 85:587-594.
Yesai r , D. W. 1984. Human disposition and some biochemical aspects of met h-
yl xant hi nes . In The Methylxanthine Beverages and Poods: Chemistry, Con-
sumption, and Health Effects, edited by G. A. Spiller. New York: Liss.
Zchme, Bi l l . 1994. Lct t cr man lets his guard down. Esquire, Dccembcr:97-102.
Zi el ke, C. L., and Ziclke, II. R. 1987. Chronic exposure to subcut aneousl y im-
pl ant ed mct hyl xaut hi nes . Biochemical Pharmacology 36:2533-2538.
Number s i n i t al i cs r ef er t o pages on whi ch i l l us t r at i ons appear.
Acet yl chol mc, and caf f ei ne, 148, 169 Addi ct i v e personality, and caf f ei ne, 177
Acctaklehyclc, 29, 34-36 Adcnosme, 127-132, 129, 138-139
AC1ID. See At t ent i on def i ci t ef f ec t s of, on blood vessels, 138
hypcr act i v i t y disorder as l i nk between alcohol and
Aci di t y, of cof f ee and t ea, 171 caf f ei ne, 183-186
Action pot ent i al , 47-48 and pr emenst r ual syndrome, 153-
Addi c t i on. See also Alcohol; 154
Alcoholism; Caf f ei ne role of, in i nt oxi cat i on, 186
to alcohol, 35-36, 88-106 and ur i nat i on, 149-150
to c a f f e i ne , 162-168, 174-177, 180 v ar i at i on in, between i ndi v i dual s ,
and mcsol i mbi c area of br ai n, 55-56 166
Index I 207
Adenosine receptor, upregul at i on,
168-169
Adenosine receptor t ypes, 138-139.
See also Adenosine
Adhesion molecules, ef f ec t s of alcohol
on, 81
Adrenaline, ef f ect s of caf f ei ne on, 134,
144, 169
Advertising, and alcoholism, 102-103
Aggression
alcohol e f f e c t s on, 78
expectancy effect s on, 78
Alcohol. See also Et hanol
absorption of, from small i nt est i ne,
33
absorption of, f r om stomach, 28-29,
32
anaerobic metabolism of, 17-18
ani mal responses to, 8889, 95 97
as ant i depr essant , 59-60
as ant i oxi dant , 66
and anxi et y, 52-54
and caf f ei ne, synergi st i c ef f ect s of,
70-71, 182-188
calories in, 16
and common cold, 73-75
congeners in, 85-86
consumpt i on of, 6
and creat i v i t y, 188-190, 194
deaths from, 165
depressant act i ons of, 38-39, 45-46,
49-50, 52-54
di st i l l at i on of, 16-17
di ur et i c propert i es of, 82-83
ef f ect s of, on brain cells, 40-41
ef f ect s of, on liver, 33, 67
ef f ec t s of, on REM sleep, 69-70
ef f ect s of, on sperm, 81
el i mi nat i on of, from body, 29, 33-
34
f er ment at i on, 9, 15-16
and hangovers, 82-86
heal t h i mpact s of, 62, 71-75
and hear t disease, 62-68
and i mmune system, 72-75
and l act i c acid, 17-18, 83-84
and memory, 50-52
metabolism of, 17-18, 34-36
and orgasm, 75-76
and pai n, 25-27
and pr egnancy, 80-82
and pr emat ur e ejacul at i on, 76
and serotonin, 59-60
and sexual response, 75-80, 86
and sleep, 68-71, 86
as solvent, 13-14, 43-45
st andar d uni t of, 13, 115-116
st i mul ant act i ons of, 38-39, 56-57,
68-69
and st i mul at i on of stomach acid,
28
and stomach, 27-29, 32
and sugar , 14-16
taste of, 21-27
t ol erance to, 69, 95-97
types of, 10-11, 43
withdrawal f r om, 84-85, 95-97
Alcohol dchydrogcnase, 29-30, 32-36,
83
Alcohol flush react i on, 35-36
Alcoholism
and adv ert i si ng, 102-103
ani mal models of, 88-89, 95-97
and controlled dri nki ng, 87
and D, receptor, 97-101
disease model of, 91-95
ef f ect s of, on sexual response, 80
and endorphms, 101
env i ronment al f act or s in, 3536, 92-
94, 100-103
genetic fact ors in, 34-36, 90-101,
105-106
hent abi l i t y pat t erns in, 92-94
in Japan, 35-36
and parent al i nfl uences, 103
prevalence of, in f emal es, 92
prevalence of, in males, 9293
and serotonin, 59-60, 100-101
and st ress, 102
t r eat ment of, 59, 100-101
t wi n studies of, 93-94
types of, 96-97
208 / Index
Al cohol - r el at ed t r af f i c acci dent s, 50,
63
Al dehyde dchydr ogenasc, 29, 34-36
Al kal oi ds , 112
Amphet ami ne, 39, 5556, 163
Anaer obi c met abol i sm
of al cohol , 15-18
by b a c f c n a in gut , 30
An f i o x i d a n l , pr oper t i es of al cohol , 66
Anxi et y
and alcohoi, ' } 254
and c a f f e i n e , 164
Apollo, t empl e of, 195
Arabi ca cof f ee. See Coffea arabica
Ar i s t ot l e, 3-4, 38-39, 45, 54
As i a ns , di f f e r e nc e s of alcohol
met abol i sm in, 34-36
Aspirin
c a f f e i n e i n, 150-152
e f f e c t s of, on al cohol met abol i sm,
32
and hangov er s, 83
Ast hma, 117-118
At hl et i c per f or mance
and alcohol, 17-18, 83-84
and c a f f e i n e , 142-146, 149
At t ent i on def i ci t hypcr act i v i t y
di sor der , and caf f ei ne, 179-180
Bach, J oha nn Sebast i an, 132-133, 135
Bal / ae, Honore de, 133
Beer, consumpt i on of, 108
Ben/ odi az epi nes, 52-54, 68, 175. See
aho Val i um
Bct a-cst radi ol , and c a f f ei ne, 153
Bl ackout s, al cohol - r el at ed, 52
Blum, Kenneth, 98-99
f i o d h i d h a r ma , 1 1 0
Bone densi t y. See Osteoporosis
Br ai n
bal ance of neur ot r a ns ni i t t er s i i r , 128
l i mbi c system of, 190-191
mul t i pl e f acet s of, 190-191
ncocor i cx of, 191-192
neur ons in, 41
pl a s t i c i t y of, 167-168
as sex organ, 77-78, 86
s t r uc t ur e of, 41-43
synapses in, 42-43
types of r ecept or s in, 58-59
v ar i at i ons in, 104
Br andy, congeners in, 85
Brewing, h i s t o r y of, 9
Buddhi s m, l i nks t o tea, 110
Cacao, 117-118
Cafe, 111
Ca f f ei ne. See also Coffee; Tea;
Tr i met hyl xan t hi ne
and acct yi chol i nc, 148
addi ct i on to, 162-168, 174-177, 180
and adcnosi ne, 127-132, 129
and a dr ena l i ne, 134
and al cohol , syner gi st i c e f f e c t s of ,
70-71, 182-188
as ant i dot e to alcohol, 183
and at hl et i c per f or mance, 142-145
average consumpt i on of, 176
and basal met abol i c rate, 146-147
bi phasi c act i on of, 131-132
and breast f eedi ng, 120-121, 164
as bronchodi l at or, 117-118, 138
and bi cycl i ng, 143-144
and cr eat i v i t y, 132-133, 188-190,
194
and defecat i on, 149-150
as depressant , 178-180
and di et i ng, 145-148
in die! pi l l s, 145-146
di scov ery of, 111-112
and eat i ng di sorders, 147
in Enl i ght enment society, 124
etymology of, 1 11
and fat , ]'44-147
and gas t r oi nt es t i nal problems, 176
ha l f - l i f e of, 70, 120, 121, 156
and headaches, 83, 172-174
heal t h ef f ec t s of, 137-139, 164
and hear t di sease, 148-149, 164,
176
and hunger , 147
and hypc r a c t i v i t y, 179-180
and i mpul si v eness, 135-136
Index I 209
i ndi v i dual v ar i at i ons i n response t o,
136
and l ear ni ng, 1 341 36
mechanism of action of, 1 1 5 , 127-
132
and memor y, 133134
and mens t r ual cycle, 120
met abol i s m of, 119-122
mol ecul ar propert i es of, 114-116,
115
and muscl e t wi t ches , 148-149
na t ur a l sources of, 107-108, 112
in nonpr escr i pt i on dr ugs, 6
and oral cont r acept i v es , 120
and ost eoporosi s, 157-158
overdose pot ent i al of, 130-131
and pai n, 151- 152
paradoxi cal ef f ect s of, 178-180
per capita consumpt i on of, 6
personal e x pe r i me n t a t i o n wi t h, 181
pest i ci dal propert i es of , 1 1 21 1 4
pot ency of, 163
and pr egnancy, 155-157
and pr emenst r ual syndrome, 152
1 55
and r ef l exes, 144
r el at i on of, to ot her addi ct i ons, 177
and r unni ng, 143
saf et y of, 164-165
and sexual response, 139-142
and sleep, 70-71, 125-128
and sper m, 141
s t andar d uni t of , 115116
subt l et y of e f f e c t s of, 163-164
as t cr at ogcn, 156
t ol erance to, 150- 151, 165, 167-171
as t r i r net hyl xant hi nc, 115, I I 5
and ur i nat i on, 149-150
wi t hdr awal f r om, 150-151, 165,
171-177
wi t hdr awal sympt oms of, 173-174
Ca f f ei ne Anonymous, 162163
Ca f f c i ms m. See Ca f f e i ne : addiction to
Ca f f er got , 150
Calories, in alcohol, 16
Cancer, and c a f f e i ne , 164
Capsai ci n, 26
Charness, Mi chael , 81
Chocolate, 117-118, 124
Cholesterol
and alcohol, 64-67
and dec a f f ei na t ed c of f ee, 158-160
Ci met i di ne, 32
Clark, Joel, 118
Coca-Cola, 108-109
Cocaine, 39-40, 55-56, 113, 163
i n sof t dr i nks , 109
Coffea arabica, 108
Coffea robusta, 108
Coffee. See also Ca f f ei ne
aci di t y of, 171
bu/ z from, 120, 171
caf f ei ne in, 115
consumpt i on of, 108, 109
cul t i v at i on of, 113
decaf f ei nat ed, 158-160
hi st or y of, 123-125, 137-138
legends of, 110-111, 139-140
t hcophyl l me in, 116-117
Coffee Cantata, 133
Coffeehouses, 140
Cognac, congeners in, 85
Colds, e f f e c t s of alcohol on, 73-75
Congeners, in alcohol, 85-86
Controlled dr i nki ng, of alcohol, 87
Cooling, ef f ect s on t ast e, 26-27
Cooper, Leon, 132
Craving, for alcohol, 54
Cr eat i v i t y
and alcohol, 188-190, 194
and c a f f ei ne, 132-133, 188-190,
194
Decaf f ei nat ed cof f ee, 158-160
consumpt i on of, 158
decaf f ei nat i on processes, 159-160
and heart disease, 1 59
Def ecat i on, ef f ect s of caf f ei ne on,
149-150
Del i r i um trcmens, 95
Depression
ef f ec t of alcohol on, 59-60
2101 Index
Depression (emit.)
as s ympt om of c a f f ei ne wi t hdr awal ,
173
Diagnostic and Statistical Manual of
Mental Disorders, 175-176
Diet pi l l s , c a f f e i n e i n, 145-146
Di mct hyl xant hi ne, 114, 117, 119
Disease model of al cohol i sm, 91-95
Di st i l l at i on, 16-17, 7 ' l , 1 11
DNA. See Genetic di f f er enc es in
alcohol pr ef er ence
Doparmne, and alcohol, 5657, 68,
97-101
and c a f f e i n e , BO, 186
D, receptor, 97-101
Dr. Pepper, 109
DSM-IV. See Diagnostic and
Statistical Manual of Mental
Disorders
Duf our , Syl v cst rc, 185, 187-188
Dumv i ddi c, Thomas, 183-184
DuPont , 188
Kat i ng di sorders, and c a f f e i n e , 147
Emet i ne, 112
Endor phi ns
and alcohol, 39, 56-58, 68, 101
and c a f f ei ne, 130
Estrogen, and c a f f e i ne , 156
Et hanol . See also Alcohol
solvent pr oper t i es of, H14, 105
st andar d dr i nk of, 13
s t r uc t ur e of, 10-14, JO, 105
Et hyl acet at e, 160
Exercise, and hangov er s, 8384. See
also At hl e t i c per f or mance
Expect ancy
ef f ect s of, on sexual response, 77-80
ef f ec t s of, on al cohol i sm, 102
Fat t y aci ds, ef f ect of c a f f e i n e on, 144-
' l 45
EAS. See Fet al Alcohol Syndr ome
Fat i gue, and c a f f e i n e us e, 173, 178-
179
Fer ment at i on, 9, 15-16
Fet al Al cohol Syndrome, 81
Fi t z ger al d, I-' . Scott, 189-190
For mal dehyde, 1 1
Eol t s, John, 6566
French paradox, 63
GABA. See Gai mna- ami nobuf yr i c acid
Gamma- ami nobut yr i c acid
ef f ect s of alcohol on, 53-54, 57, 68
ef f ect s of caf f ei ne on, 169
rel at i on of, to v ar i at i on in response
to alcohol, 97
Genet i c di f f er ences i n alcohol
pr ef er enc e, 90-101, 105-106
Gcrasch, Sylvia, 142
Gi n, congeners i n, 85
Glucose, and al cohol , 15
Gl u t a ma t e recept ors, act i ons of
alcohol on, 49-52
Glycol, 10
Gorbachev, Mi khai l , 8
Greene, Robert, 127
l l a h n e ma n n , Samuel, 125
l l ai r - of - t he- dog, 84-85
[l al ci on, 68
Hangov ers, 82-86
and as pi r i n, 83
ki nds of, 85-86
HDT..S. See Hi gh-densi t y l i popr ot ci ns
Headaches
a l l ev i a t i on of , by c a f f e i n e , 150152
pos t oper at i v e, 172
as r esul t of c a f f ei ne wi t hdr awal ,
172-174
as resul t of hangover, 8283
Hea r t disease
and al cohol , 62-68, 86
and c a f f ei ne, 148-149, 164, 176
and decaf f ei nat ed cof f ee, 158-160
Hemi ngway, Ernest , 188-190
Hemogl obi n, 1 2
He r oi n, 55, 57, 168, 171
Index I 21 I
Hi gh-densi t y l i popr ot ci ns, and al cohol ,
64
Howell, James, 124
Hunger, and caf f ei ne, 147
Hyper aet i v i t y. See At t ent i on defi ci t
hyper act i v i t y disorder
I mmune syst em, ef f ect s of alcohol on,
72-75
Impul si v eness, e f f e c t s on c a f f e i ne ,
135-136
I nsomni a, and alcohol, 68
I nt oxi cat i on
and adenosme, 186
and caf f ei ne, 186
and congeners, 85-86
and dopamme, ' 3657, 186
euphoric ef f ect s of, 56
gender di f f er ences in, 5031
i on channel t heor y of, 45, 104-105
legal def i ni t i on of, 41
lipid t heory of, 43-45
Ion channel s
closed, 24
and i nt oxi cat i on, 45, 104-105
and nerve-cell firing, 46-47
opeir, 25
st r uct ur e of, 22-25
Jensen, Crethe, 41
Johnson, George, 132
Johnson, Vi r gi ni a, 76
Kal di , 110-111
Kant, I mmanuel , 133, 135
Kola nut s, 108-109
Kor sakof f ' s syndrome, 72
Lact i c acid
as byproduct of alcohol met abol i sm,
17-18, 83-84
and hangov ers, 8384
LDLs. See Low-density l i poprot ei ns
Legendre, Rene, 126
Let t er man, David, 188
Li nnaeus, Carolus, 111
Li pi d t heor y of alcohol' s act i on, 43-45
Li ver, 33-34
as si t e of caf f ei ne met abol i sm, 119
120
di seases of, caused by alcohol, 33,
67, 72
Long-term i nhi bi t i on, 50
Long-term pot ent i at i on, 50-51, 133-
134
Low-density l i popr ot ei ns
and alcohol, 65-66
and decaf f ei nat ed cof f ee, 158-160
LSD, 40, 116, 163
LTP. See Long-term pot ent i at i on
Lucretius, Titus, 103-104
MacKi nt osh, James, 133
Mast er s, William, 76
Mat e, 108
Maxwell House, 118
McCl earn, Gerald, 89
Mellow Yellow, 109
Memory
ef f ect s of alcohol on, 50-52
e f f e c t s of caf f ei ne on, 133-134
t heor y of, 50
Menst r ual cycle, and c a f f ei ne, 120
Mescaline, 112
Mesolimbic area, of brai n, 55-57
Metabolism, set poi nt of, 167
Methanol, 11, 85-86
Methylcne chloride, 160
Met hyl groups, 114, 117, 119-120
Met hyl xant hmcs, 114-120, 133, 149
and memory, 133-134
Meyer, H. , 43'
Meycr hof , Kmbdcn ()., 15
Mice
as research subj ect s, 88-90
al cohol - pr ef er r i ng, 89-90, 95-96
Mi chel ct , Jul es, 140
2121 Index
i VI j gr ani e headaches, and c a f f e i n e , Pcmbcrt on, John, 109
1 5 0 - 1 5 1 Pepsi -Col a, 108-109
Mi nd, as s o c i e t y , 1 9 0 1 9 1 Phenol s, in wi ne, 66
Moder a t i on Phcnyl pr opanol ami nc, 146
in al cohol c o n s u mpt i o n , 86-87 Phi l l i s, John W., 153-154
d e f i n i t i o n of, in d r i n k i n g , 87 Phosphodi cst cr asc, inhibition by
Mor phi ne, 12, 1 2, 57, 1 1 2- 1 1 3, 175 c a f f e i n e , 131
Mo u n t a i n Dew, 109 Pi cron, Henr i , 126
Pl easur e cent er . See Reward Cent er
Naegel i - Moody, Mar s ha, 163 PMS. See Pr emens t r ual syndrome
Na t h a i r s o n , J a mes , 112- 113 Pol ymodal pa i n f i ber s, 25-27
Na t u r a l k i l l e r c e i l s , e f f e c t s of al cohol Pr egnancy
on, 72-7? and al cohol , 63, 80-82
Nausea, as . sympt om of c a f f e i n e and c a f f e i ne , 155-157, 164
wi t h d r a wa l , 173 Pr e ma l u r e e j a c u l a t i o n , ef f ec t s of
Nc u r o n a l f i r i n g , 46-47 alcohol on, 76
Ne u r o n s Pr emens t r ua l syndr ome, and c a f f e i ne ,
f unc t i on of, 46-49, 4< S 152- 155
number of, i n br ai n, 41 Proctor, Wi l l i am, 183-184
number of , lost i n l i f e t i me , 41 Progesterone, and c a f f e i ne , 153
s t r uct ur e of, 42 Pr ohi bi t i on, 71
Ne u r o t r a n s mi l t e r s , 47-50 Proof, d e f i n i t i o n of, 17
Ni cot i ne, 55, 112- 113, 164. See a/ so Pro/ ac, and al cohol , 59-60, 100-101.
Tobacco Sue also Ser ot oni n
and c a f f e i n e me t a b o l i s m, 1 2 I Psychoact i v e dr ugs, pl a nt sources of,
po pu l a r i t y of, 6 1 1 3
source of, I 07-408
NMDA r ecept or , 51-52
- . , i , ,- , 01, < < One r c e t i n, in wi ne, 66
Nobel, Er nest , 98-99
Nor epi nephr i ne. See Adr e na l i ne
Nnng, Shcn, 1 1 0 Ra mt i di nc , 32
Rat s, as research s ubj ect s , 88-89
Olds, J ames , 55 Rebound e f f e c t , t o alcohol
Ol i v e oil, e xpe r i me nt s wi t h, 43 consumpt i on, 69-71, 8486
Ol ympi cs, and c a f f e i n e , 142-143, 145 Red wine
Opi at es , act i ons of, 57-58, 17' ) congeners in, 85-86
Opi um, 39, 56-57 and heart di sease, 63-68, 86
Oral cont r acept i v es, and caf f ei ne, 120 Ref l exes , and c a f f e i n e , 144
Organi c c u l t i v a t i o n of cof f ee and l e a , Regcst ci n, Quent i n, 178
113 REM sl eep, ef f ec t s of al cohol on, 69-
Ost eoporosi s, ant ! c a f f e i n e , 1 ^7-4 58, 70
164 Res t or i l , 68
Ov er t oi l , E., 43 Reward cent er, of br ai n, 54-58, 98,
191-192
Pa r a xa nt hi ne, 119- 120 Ri t a l i n , 179-180
Pat ent medi ci nes, and al cohol , 71- 72 Robus t a cof f ee. See Coffea robusta
Index I 213
Rodgers, Dav i d, 89
Roosevelt, Theodore, 118
Rossignol, Annet t e Mac-Kay, 152- 155
Rousseau, Jean-Jacques, 133
Rut gers U ni v er s i t y, 75, 77
Rut i n, in wi ne, 66
Scotch whi s ky, 21 , 24
congeners in, 85
Serot oni n
and alcohol, 59-60, 100-101
and caf f ei ne, 169
Set pomt , caloric, 167
Sexual response
and al cohol , 75-80, 86
and caf f ei ne, B9-142
and chronic dri nki ng, 80
ef f ec t s of expect ancy on, 77-80
Shakespeare, Wi l l i am, 75, 82
Sleep
e f f e c t s of alcohol on, 68-71, 86
ef f ect s of caf f ei ne on, 70-71, 125-
128, 132, 178-179
Smoki ng
e f f e c t s of, on caf f ei ne met abol i sm,
121-122
ef f ect s of, on i mmune syst em, 73
75
Socrates, 112
Soft dr i nks
c a f f e i ne in, 108-109, 115
consumpt i on of, 108
Soviet Union, 8
Sperm
e f f e c t s of alcohol on, 81
e f f e c t s of c a f f e i n e on, 14]
St andar d dr i nk, of al cohol , 13, 79
number of et hanol mol ecul es i n,
13
Steinbeck, John, 189-190
Stomach, as si t e of alcohol absor pt i on,
27-29, 32
St omach aci d, 27-28, 171
St r a i n, F,ric, 176
Stress and al cohol i sm, 102
St r ychni ne, 112
Substance abuse, 175, 180
Subst ance dependence, 175-176, 180.
See also Addi ct i on: to c a f f ei ne
Sugar , as origin of alcohol, 14-16
Sumcr i ans , and brewing, 9
Synapses
' f unct i on of, 47-48, 48
number of, in brai n, 42
Tea. See also Caf f ei ne
a c i di t y of , 171
c a f f e i n e i n, 115
and c r ea t i v i t y, 132
cul t i v at i on of , 113
heal t h e f f e c t s of, 137-138
i n t r o d u c t i o n of, t o Europe, 124,
137-138
legends about, 110, 139-140
and pr emens t r ual syndrome, 153
t heophyl l i ne in, 116-117
worl dwi de consumption of, 108
Testosterone
ef f ect s of, on alcohol, 78-80
and femal e sexual response, 78-79
Tet rahydrocannabi nol , 12, 12, 113
TIIC. See Tet rahydrocannabi nol
Thcobromine, 117-120
Theophylline, 116-117, 117, 119-120,
184
Tibet, use of tea in, 143
Tobacco, deaths f r om, 165
Tolerance
to alcohol, 69, 95-97
to c a f f e i ne , 69-71, 165, 167-171,
175, 180
Tongue, 22-27
ion channel s i n, 2225
t ast e buds in, 22
Tr af f i c accidents, and alcohol, 50, 63
Tr i met hyl xant hi nc, 114-115, 115. See
also Caf f ei ne-
Twin st udi es, of alcoholism, 93-94
Upr cgnl at i on of adenosme recept ors,
168-169
214/ I n d e x
U ri nati on
ef f ec ts of caf f ei ne on, 149-150
ef f ec ts of alcohol on, 82 83
V al i ui n, 39, 52-54, 68, 102, 175
V an Lecmven, Storm, .144
V arrati on, i ndi vi dual
i n response to alcohol, 103-106
i n response l o caf f ei ne, 136,
165-168, 173-174, 176-177,
179-181
V odka, congeners i n, 85
V ol tai re ( K rangoi s-i V l ane A rouct),
133
V omi ti ng, and al cohol , 2.8, 84
Whi te wi ne
and heart di sease, 66
congeners i n, 85
Whi tman, Walt, 190-191
Wi ne. See also Red wi ne; Whi te wi ne
alcohol content of , 16
and heal th, 62-68
Wi thdrawal
f rom alcohol, 84-85, 95
f rom caf f ei ne, 150-151, 171-177,
180
Women. See also Menstrual cycle;
Osteoporosi s; Pregnancy
ef f ects of alcohol on testosterone
levels i n, 78-79
and moderate alcohol consumpti on,
30-31
speed of i ntoxi cati on of , 30-31
Wood al cohol . See Methanol
X anax, 68
X anthme, 114, 114, 117, 120
Y east, and f ermentati on, 15 17
7,cn, and tea, 110