M.

C STERLING, UNIVBRSITY OF CALIFORNIA

RECENT DISCUSSIONS OF ETERNAL RECURRENCE:

SOME CRITICAL COMMENTS

Chapter 7 of Arthur Danto's Nietzsche äs Philosopher is devoted, in

part* to a consideration of Nietzsdie's version of eternal recurrence.
Although DantoV discussion of eternal recurrence strikes me äs unduly
Brief (especially in view of the tremendous importance which Nietzsche
assigned to it), I think it would be appropriate to consider it here, for he
does touch upon some significant issues. I shall begin by presenting a gen-
eral sununary of Danto's treatment of eternal recurrence; after this, I
shall try to make some critical comments about it.
Danto first cites two passages from Nietzsche that deal with the eter-
nal recurrence idea. For convenient reference later on, let us designate the
first passage äs A, and the second äs B:
If the world dare be thought of äs a determinate magnitude of power,
and a determinate number of power centcrs — and every other idea
is incteterminate and hence unusable — it follows that is has mn through
a calculable number of combmations in the great dice game of its
cxistence. In an infinite time, every possible combination would some-
time have been attained: more, each would have becn attained an
infinity of timet, And thcn, between eadi combination and its ncxt
repctition, all the remaining combinations must then be run through,
and cach of diese combinations determines the wholc sequencc of com-
binaüons, so that a whole cycle of absolutely idcntical sequenccs rcsults.
The world is a cycle whidi has already infinitely rcpeated its elf, and
plays its play in infinitem.1
The total amount of energy fAU~Kraft] is limited, not "infinite.* Let
us beware of such conceptual excesses: Consequcntly, the number of
statcs /Lagen], combinations, dbanges, and evolutions [Entwicklungen]
of this energy is tremcndously great and practically immeasurable, but
in any case finitc and not infinite. But tbe time through whidi this total
energy works is infinite, That mcans tbe energy is forever the samc and
forcvcr active, An infinity has already passed away bcforc this prcscnt
morr.cnt. Tbat means that all possible dcvelopmcnts must have taken

1
SA III, p. 7C4,
262 M, C Sterling

place already. Consequently, the present development is a repetitton,

and thus also that whidi gave rise to it, and that which arises from it,
and so backward and forward again! Insofar äs the totality of states
of energy [die Gesamtlage aller Kräfte] always recurs, everything has
happened innumerable times.2
At this point, Danto claims that A is "exceedingly garbled," and that B is
a "more nearly adequate account."
Next, Danto lists three propositions, presumably isolated from B, and
proposes to add any further premises "required to get the conclusion Nietz-
sche needs." The three propositiöns are:
(1) The sum-totai of energy in the universe is finite.
(2) The number of states [Lagen] of energy is f inite.
(3) Energy is conserved.
With respect to these propositiöns, he maintains the following: (ä) They
are logically independent; (b) Nietzsche believed (2) to be entailed by (1);
(c) Nietzsche was mistaken in believing (2) to be entailed by (1); and (d)
taken by themselves, (1), (2), and (3) do not entail eternal recurrence. I
shall come back to (d) a bit later; for the moment, let us consider what
Danto has to say about (a), (b), and (c).
(a), (b), and (c) äbove are logically interconnected; for (a) and (b),
taken together, imply (c), Now, the truth of (b) is strongly suggested by
the cited passages, A .and B, so that, to establish (c), Danto might con-
centrate on showing (a) to be credible. This, in fact, is the course which
Danto follows. He i? primarily concerned to demonstrate that propositiöns
(1) and (2) are logically independent of one another, and, consequently,
that Nietzsche is wrong in regarding (2) äs implied by (1); so, he initially
ignores proposition (3). His diief contention is that because there is a
"wholly natural Interpretation" of 'Lage* in which (1) would be true and
(2) would be false, it cannot be the case that (2) is entailed by (1).
To illustrate his point, he of fers the example of a conservative energy
System with a f inite amount of energy:
Imagine a conservative energy System with a f inite amount of energy.
For the sake of simplicity, let us say that the amount is equal to a
f inite number, 6. Süppose some of the energy is kinetic. Suppose again
that äs the kinetic energy increases, the potential energy decreases; the
rate is sudx that the latter approadies 0 äs the förmer approadies 6.
These limits could be approached indefinitely without being readied.
Now let Lage mean "the amount of kinetic energy plus the amount of
Potential energy at any given instant.* There could be an infinite
number of Lagen, then, and no Lage need »recur. On sudi a model,
2
GA XII, p. 51—52. .
 Recent Discussions of Eternal Recurrence 263

(1) and (3) would be tfue, and (2) would be false. So (2) is indepcndent
of (1) and (3).«
At this point, Danto begins to list four additional propositions whidi
he beiieves are required to deduce eternal recurrence; more specifically, he
thinks that four additional premises are needed in order to deduce that no
Lage (i. e,, state of the universe) has had a first occurrence and, hence, that
each Lage has occurred an infinite number of times. The additional pro-
positions are äs follows:
(4) Tnneis infinite.
(5) Energy has infinite duration.
(6) Change is eternal.
(7) Prineiple of Sufficient Reason.
Thus, Danto beiieves that a total of seven propositions is required in order
to derive a conclusion awhich is pretty much Nietzsche's theory."
The gist of Danto's discussion of eternal recurrence has just been
presented; let us now take a closer look at what he is saying. It appears
that Danto is making two major claims. First, he contends that it is not
the case, äs Nietzsche beiieves, that a finite amount of energy (Kraft) im-
plies a finite number of states of energy (Lagen der Kraft). Second, he
maintains that seven lögically independent propositions are required to
deduce Nietzsche'* doctrine of eternal recurrence. I shall consider these
two claims in turn.
To begin with, the truth of Danto's first claim (i. e., that proposition
(1) does not entail proposition (2)) depends on two things; specifically, i t
depends on (a) what is meant by camount of energy/ and (b) what is meant
by 'state of energy/ Recognizing this, Danto offers, for both expressions, a
sense whidi he thinks establishes his claim; however, to refute Nietzsche's
Suggestion that finite energy implies a finite number of states of energy,
the irnportant qtiestion to ask is: What did Nietzsche mean by these phrases?
Danto appears to believe that what Nietzsdbe meant by these expres-
sions is very unclear and, hence, that one is at liberty to advance a
e
wholly narural Interpretation* of them. But the fact remains, I think, that
unless this "whoüy natural Interpretation1* of 'amount of energy' and
'state of energy* is actuaüy what Nietzsche meant by these expressions,
Danto cannot show Nietzsche to be wrong by establishtng that, on this
a
whoily natural Interpretation,^ entailmcnt betwcen (1) and (2) collapses.
Consequently, I shall first try to determine what Danto means by famount
of eocrg>^ and 'state of energy/ and then $ee if this has the consequence, äs
he bclicvcs, that propositions (1) and (2) are lögically independent» Sub-
9
Danto, Arthur, Nktzichc äs Phifaiopbcr {New York and London: Macmülan Co.,
1965), p. 206.
264 M. C. Sterling

sequently, I shall attempt to get some insight into what Nietzsche nnght
have meant by 'amount of energy' and 'state of energy,' in intimating that
(2) is entailed by (1).
Consider again Danto's exampel of the "conservative energy system."
Now, it seems to me that there is something seriously wrong with this
example and with the conclusions which Danto draws from it. For one
thing, instead of clarifying the expressions 'amount of energy' and 'state of
energy/ to my mind at least, it obsCures them. In considering Danto's
example, however, I get the feeling he has vaguely in mind the Standard
scientific meanings of these expressions. But, in any eveiit, he is suggesting
that, given the senses of 'amount of energy' and 'state of energy' implicit
in his example, entailment between propojitions (1) and (2) collapses. I
contend, however, that regardless of the exact senses of these expressions
that Danto might have in mind, this claim is false; moreover, it is false
in virtue of the very assumptions of his example.
For, recall that Danto says "let Lage rnean the amount of kinetic
energy plus the amount of potential energy at any given instant." If we do
äs Danto is requesting here, then, on his further assumptions that: (1) the
potential energy approaches 0 äs the kinetic energy approadies 6, and (2)
that the total amount of energy (equal to 6) is conserved, it does not follow
that "there could be an infinite number of Lagen, then, and no Lage need
everrecur."
In fact, there could only be one Lage, namely 6; for if Lage means
the amount of kinetic plus potential energy, it must always be equal to 6.
For example, if the kinetic energy were 4, then the potential energy would
be 2, and, consequently, the Lage would be 6; again, if the kinitic energy
were 5, the potential energy would be l, and the Lage would still be 6, and
so on. Incidentally, this strikes me äs being a very odd way of speaking;
but, then, Danto's whole example of the conservative energy System strikes
me äs odd (insofar äs its relevance to Nietzsche is concerned), and this
manner of speaking is implied by it. This leads me to süspect that there
may be something here äbout whidi Danto is slightly confused. (To get the
result he wanted, i. e., that "there could be an infinite number of Lagen,
then, and no Lage need ever recur," Danto should have defined Lage, not
äs the amount of kinetic energy plus potential energy, but rather äs either:
(1) the value of kinetic energy at any given instant, or (2) the value of
potential energy at any given instant, The value of kinetic energy, for
example, might at a given instant be anywhere from 0 through 6, and
since there i§ an infinity of values between 0 and 6, if we define Lage äs the
value of kinetic energy at any given instant, there would be an infinite
number of Lagen and no Lage would have to recur. The same cpnsiderations
 Recent Discussions of Eternal Recurrcncc 265

would apply if we defined Lage äs the value of potential energy ät any

given instant; but if we say, äs Danto does, that Lage means the araount
of kinetic energy plus potential energy, then the Lage must always be 6,
and Danto cannot get the result that he wants.)
But what did Nietzsche mean by €Mass der All-Kraft* and 'Lage der
All^Kraft) i.e., according to Danto's translation, 'amount of energy' and
'state of energy?* Although Nietzsche had the habit of taking certain
liberties with language, and of exploiting the multiple connotations of
words for his own ends, I think he meant something fairly definite by
these expressions. To get at Nietzsche's meaning, I think it is important to
inquire about his ontologicai belief s (I am, sympathetically, assuming that
Nietzsche's conception of "what there is* was such äs would sanction his
Suggestion that proposition (1) entails proposition (2)). Accordingly, I shall
now sketch the broad outlines of an ontology which, I believe, allows this
inference, and shall then compare it with Nietzsche's own ontologicai
pronouncements,
Imagine the following pluralistic ontology: The universe, let us say,
consists of a finite, though enormous number of ontologicai units. By Ont-
ologicai unit' may be understood a basic unit of *what there is/' i. e., a
simple entity. Let us assume that these ontologicai units are diaracterized
by an all-or-none sort of interaction, By call-or-none sort of interaction* I
mean an interaction such that, at any given instant, eadi ontologicai unit
is either interacting in toto, or it is not interacting. In other words, there
can be no dcgree$ of this relation of interaction; an all-or-none relation of
this son either obtains completely, or it does not obtain at all. For example,
the relation 'toudbing' has an all-or-none diaracter. Thus, two objects,
and y, must be either toudiing completely> or not touching at all; there can
be no degrees of this relation, for it is an either-or, all-or-none kind of
relation. On the other hand, the relation, cbeing far from/ lacks an all-
or-none property. I can say that an object, x> is far from an object, y;
but sincc there can be degrees of this relation, I can also say that a third
object, z, is farther from y than is. Thus, with this all-or-none sort of
interaction, there are only the following four possibilitics: A given onto-
logicai unit must either be interacting with (1) any orte of the other units;
(2) any wbszt of the other units; (3) all of the other units; (4) none of
the other units. Thus, there are no other states of these ontologicai units
ovcr and abovc their interacting or not interacting; there is no basis, for
example, for defimng scvcral additional states under the one rubric 'non-
mtcracüve Diätes*; for, only onc state falls under this rubric (i.e., (4)
abovc).
266 M. C. Sterling

Now, if N Stands for the number of ontölogical units, then the num~
ber of states-of-interaction of these ontölogical units, taken r at a time, is
N!
(N-r)tr! * Thus, if 'amount of energy* is given the same meaning äs 'num-
ber of ontölogical units,' and if 'state of energy1 means the same äs 'state-
of-interaction of these ontölogical units,' then a finite amount of energy
implies a finite number of states of energy; for, on the assumption of an
all-or-none sort of interactioii, a finite number of ontölogical units entails
a finite number of states-of-interaction of these units. What is crucial to get
this entailment is the assumption of what I am calling 'all-or-none inter-
action.' More specifically, since all-or-none interaction fixes the number of
N!
states-of-interaction at />j. r \t r t 9 when the number of units is N, and since
. . . N !
the finitude of N implies the finitude of (N-rjiri, it follows that a finite
number of units entails a finite number of states-of-interaction.
This is at least one of the things that Nietzsche realized, I think, and
was insisting on when he rejected the idea of infinite space äs inimical to
his doctrine of eternal recurrence; for the assumption of infinite space
would render what I call call-or-none interaction' mudi less plausible. Con-
sider this Statement from Nietzsche's Nachlaß:
From a System of determinate forces [Krfifte], i. e., from a measurably
definite force [meßbar sicheren KraftJ no inrmmerability of states can
result. Only oft the false assumption of an infinite space, in whidi the
force [die Kraft] evaporates so to speak, is the last state an unprqduc-
tive one, a dead one.4
But now, to get to my next point, how does the ontology delineated
above compare with Nietzsdie's own? Walter Kaufmann has referred to
Nietzsdie's ontology äs "monadological pluralism"5; Kaufmanns expression
is very descriptive, in my estimation, so I shall adopt it for my own use.
According to Nietzsche's monadological pliiralism, the universe is
made up of a tremendous, but still finite number of what he called "force-
centers" or "will-points." Now, Nietzsche's writings (e. g., both A and B
above) suggest that these force-centers may enter into states pf combina-
tion,6 and, by extrapolation, we can assume they may also be in a state of

4
GA, op. cit.y p. 55 (my translation).
5
Kaufmann, Walter, Nietzsche: Philosophen ?5yc&o/og/5i, Antichrist (New York: Vint-
age Books, 1968), p. 243.
6
In view oiF Nietzsche's belief that the force-centers afe not to be thought öf äs material
atoms moving in empty space, it is stränge that he would use the term 'combination'
in talking about their interrelationships. ior, Ccpmbinati6n'/ roughly speaking, is syn-
onymous with 'juxtaposition/ whidi presupposes spatial location. I seriously doubt,
therefore, that Nietzsche inteiided to suggest that the force-cejiters, literally speaking,
 Reccnt Discussions of Etcrnal Recurrence 267

dissolution. This relation between the force-centers, moreover, may be

thought of äs an all-or-none sort of relation in the sense that, at any given
instant, a force-center is either combined completely, or it is not combined
at all; this all-or-none property of the relation, 'being combined/ is, I
think, implicit in its very definition. Thus, there are only the following
four possibilities in connection with the combining of force-centers: A given
force-center must be combined with: (1) any one of the other force-centers,
(2) any subset of the other force-centers, (3) all of the other force-centers,
or (4) none of the other force-centers. According to this, there would be no
other states of these force-centers over and above their being either com-
bined or not combined. And, in this case, Nietzsche's rejection of infinite
space, precludes the possibility of there being an infinite number of
additional states of these force-centers definable ander the one heading
f
non-combined states'; so, only one state can fall under this heading.
If, then, 'amount of energy* means the same äs cnumber of force-
centers/ and if 'state of energy* has the same meaning äs 'number of states-
of-combination of these force-centers/ we may logicaliy infer a finite num-
ber of states of energy froni a finite amount of energy. For, in this case also,
the all-or-none property of the relation between force-centers delimits the
N!
number of states-of-combination at (N-rj!r!> when the number of force-
N1
centers is N. And since the finitude of N implies the finitude of (N_ r )i r j>
the number of states-of-combination will be finite when the number of
force-centers is finite.
Contrary to Danto, therefore, I think that Nietzsdie, if his intended
senses of 'amount of energy* and 'state of energy1 coincide with the above
analysis, is not mistaken in supposing that proposition (1) entails proposi-
tion (2).
Danto's second major contenrion was that a total of seven propositions
is required to deduce Nietzsche's doctrine of eternal recurrence. It seems
to me, however, that Danto's propositions (1), (6), and (7) are sufficient
for this purpose. (1) says that there is a finite amount of energy, and (if
we assume the above analysis to be correct) it also implies that there is a
finite number of state of energy. Let m follow Danto's example and,
hence, suppose there to be thrce such states of energy, namely A, B, and C.
Now, (6) says that dbangc is etcraal, and by this is meant that the alterna-
tioa of these states of energy had no beginnbg and will never cease, Thus,

eoter toto combinatioa5, His ese of tlu» tenn 'combiÄÄtion,' I think, $hoo!d bc takcn
metaphoricaify; moreover, it h posublc that hc feit this particular mctoaym to bc
cspcdally suttäbic* because of a convictioa that the ''real" relation that obtaim
fcwrce-CTÄter$ i$f like that of liural combinatfon« an a!J-or-nonc rcbüon.
268 M. C Sterling

ncither A, B, nor C can be a terminal state; another state must always

follow.
Danto suggests, however, that we might imagine "an infinity of
alternations... A-B-A-B-A-B-A ..., and at a new cut-off point C occurs,"
In this way, he says, C would have had a first occurrence and this would be
inconsistent with eternal recurrence; so, he claims we must add (7), the
Principle of Sufficient Reason. Personally, find it hard to see how C
would ever be able to occur at all, if "an infinity of alternations* of A and
B must occur prior to Cs occurrence. How would this infinity of alter-
nations ever be completed so that C could occur?
I submit that C could never occur under these circumstances and,
consequently, that Danto's reason for thinkMg that (7) is reqüired is ille-
gitimate. Nonetheless, I do believe that (7) is needed, for a different rea-
son; specifically, Nietzsche's Version of eternal recurrence assumes that
the order in which the states of energy succeed one another is invariant,
i. e., always the same. The Principle of Sufficient Reason, then, is reqüired
for the purpose of eternally f ixing this order of the energy-states.
Now, then, assuming that (2) is implicit in (1), Dantö's propositions
(1), (6), and (7) imply the following: The universe is diäracterized by a
finite number of eternally-alternating total-states (propositions (1) and
(6)), and the order-of-alternation of these total-states is invariable (proposi-
tion (7)). This, it seems tö me, just is Nietzsdie's universe of eternal re-
currence; if so, then, Danto notwithstanding, (1), (6), and (7) are sufficient
to entail Nietzsche's doctrine.

II
Ivan Soll has attempted to critically evaluate Nietzsdie's doctrine7 of
eternal recurrence, and has paid special attention to what he thiiiks Nietz-
sche regarded äs the most important thing about the doctrine, namely the
alleged psydiological consequences of believing it. Early in his article, Soll
cites the following passage from Nietzsdie to illustrate the degree to which
the latter emphasized this aspect of his doctrine:
Th'e greatest stress. What if one day or night a deinen were to sneak
after you into.your loneliest loneliness and say to you, "This life äs you
now live it and have lived it, you will have to live once more and
innumerable times more; and there will be nothiiig new in it, but every
pain and every joy and every thought and sigh and every thing im-
measurably small or great in your life must return to you — all in the
same succession and sequence — even this Spider and this moonlight
between the trees, and even this moment and I myself. The eternal
7
Solomon, Robert C., Nictzsdw: A Collectlon of Critical Essays (New York: Andior
Books, 1973). 322—342.
 Recent Discussions of Eternal Recurrence 269

hourglass of existence is turned over and over, and you with it, a dust
gram of dust/ Would you not throw yourself down and gnash your
teedi and curse the demon who spoke thus? Or have you once ex-
perienced a tremendous moment when you would have answcred him,
*You are a god, and never have I heard anything more godly." If this
thought were to gain possession of you, it would transform you, äs you
are, or perhaps crush you. The question in each and every thing, "Do
you want this once more and innumerable times more?** would weigh
upon your actions äs the greatest stress. Or how well disposed would
you have to become to yourself and to life to crave nothing more
fervently than this ultimate eternal confirmation and seal?8
This passage does seem to suggest that Nietzsche believed certain pro-
found psychological consequences would follow upon one's acceptance of
eternal recurrence; and although it is debatable whether or not Nietzsche
actually regarded this äs the most important thing about his doctrine, it is
nonetheless not inconceivable that he did so. In any event, it is precisely
the supposition that believing in eternal recurrence has profound psycholog-
ical consequences, i. e., that it weighs upon one's actions "äs the greatest
stress,* that Soll wants to call in question.
The gist of Soll's objection to Nietzsche's notion of eternal recurrence
is embodied in the following quotation:
Contrary to Nietzsche, however, I believe the prospect of the infinite
repetition of the pleasure or pain of one's present life entailed by the
doctrine of eternal recurrence should actually be a matter of complete
indifference. To see why this somewhat counterintuitive conclusion is
the case, one shouid consider the nature of the identity between oneself
in the present cycle and oneself in any other cycle. One's recurrence in
another cycle i$ perfectly identical with one's present seif in that the
physical and psychological career, the experiences and position in the
general history of the cycle of each recurrence of the person are quali-
tatively identical.
Since for this kind of identity to hold the mental lives of eadb of the
rectirrcnces of a person must be exactly the same, there can be no
accurrmlation of experience frora one recurrence to the next. A person
can have no direct memories of earlter recurrences. If hc did, the incre-
ment of hb mental life would make him different from his predeccssors
and hence not an identical recurrence of them....
I want to suggest that only if the recurrences of one's Jife were Hnkcd
to cme*s present life äs different states of consciousness are united in one
consdouimess should one have any personal concern about the prospect
of one's present pleasur« or pains recurring. If I am concerned about
the possibility of suffering pain later in this life, it is because those futtire
iUtes of consciousness in whtdb ! might suffer pain are united with rny
present state of conscioumess. And this unity depends, at least in part

* GA VI, p. 29!—9i
270 M. C. Sterling

and in somc complex manner, upon there being linkages of memory

among these states of consciousness. Since I am not connected to my
recurrence in another cycle by being united with him in one conscious-
ness through links of memory, I am not identical to him in the sense
necessary for me to view bis suffering pain tantamount to my suffering
further pain. Only by inappropriately construing the suffering of some
future recurrence on the model of suffering later in this life does the
question of the eternal recurrence of one's pain weigh upon one with
"the greatest stress." ...
Since Nietzsche believed that the psychological consequences of the
doctrine of eternal recurrence for those who consider it at least possible
are more important than the demonstration of its truth, I have concen-
trated on those consequences. My general conclusion is that what appears
to be the momentous human import pf the doctrine is negated by its
supra-historical diaracter, which removes its consequences beyond the
possible limits of individual human consciousness and the cumulative
historical consciousness of mankind^9.
On the surface, Soll's reasoning seems quite appealing; nonetheless,
I believe his argument is only superficially plausible, and that, when
examined more closely, it exhibits serious shOrtcomings. In an effort to
show why this is true, I should like, first of all, to clearly state what I take
Soll's major contention to be. The principal claim that Soll is making is
that future joy or suffering in a recurrence of my present life should be a
matter of complete indifference to me. Now; granting this, the question
arises: What is it that distinguishes future joy or suffering in a recurrence
of my present life f rom future joy or suffering in my present lije, in virtüe
of which I ought to be indifferent aböut the former, but concerned about
the latter?
Soll's reply to the above question would be that future joy or suffering
in a recurrence of my present life is not connected, "through links of mem-
ory," to the sequence of experiences which eonstitutes my present life,
whereas future joy or suffering in.my present life is connected, "through
links of memory," to the other experiences of my present life. Thus, Soll
believes that concern or indifference is justified äs a function of the pres-
ence or absence, respectively, of a continuity öf memory. Accordingly, bis
main thesis becomes: Becäuse there is no continuity of memory between the
experiences of my present life and the experiences of a future recurrence
of my present life, future joy or suffering in a recurrence of my present
life should be a matter of complete indifference to me.
To make explicit the logical strücture of Soll's reasoning, let us recast
it in the form of "modus ponens":

9
Solomon, op. «"f., 339-T-342,
 Recent Discussions of Eternal Rccurrencc 271

Premise 1: If there is no continuity of memory between my present

life and a future recurrence of my present life, then future
joy or suffering in a recurrence of my present life should
be a matter of complete indifferente to me.
Premise 2: There is no continuity of memory between my present life
and a future recurrence of my present life.
Conclusion: Future joy or suffering in a recurrence of my present life
should be a matter of complete indifference to me.
Now, it is premise l of SolPs argument on which I should like to
focus attention, for it strikes me äs highly questionable. It presupposes
that there is some way in which the justification of concern or indifference
depends on the process of remembering; but, what exactly does it mean
to say that the justification of concern or indifference depends on the
process of remembering? It means, I think, that if I shall not be able to
remember the joy or suffering which I experience now, when such joy or
suffering recurs in the future, then the fact that it will recur in the future
should be a matter of complete indifference to me. Now, with respect to
this, I should like to raise two questions. First, simply, is the Statement
true? Secondly, if, äs I shall maintain, it is not true, what sorts of con-
siderations might conceivably have led Soll to believe it true?
In responding to the first question, let us consider the sort of Situation
in which Soll would concede that future joy or suffering should be a
matter of concern to us, i. e., the sort in which we should not be indifferent
about such joy or suffering. The kind of Situation I am referring to is that
in which future joy or suffering is to occur later on in this present life.
Imagine that I am suffering intense pain at the present moment, and that
this suffering will shortly be terminated; imagine, furthermore, that, for
some reason or other, I believe that I am going to have to re-experience
this episode of pain next week. Now, I would by anything but indifferent
about this future pain; in fact, I would be quite concerned about it. More-
over, I submit that I would be concerned about it, not because at that
future date I shall be abie to remember having experienced the same pain
a week previously; but rather because at this present moment I am able to
jorcsee or to anticipate that I shall suff er such pain a week f rom now.
Perhaps something similar is true in the case of future suffering in a
recurrence of rny presenc life. If so, then it follows that what justifies my
bcing concerned or indifferent about future süffering in a recurrence *of
my pn^ent life is the fact of my being able or tmable to anticipate that
suffering during my present life, and not my abtiity or inability to re~
meniber the suffering of my present life at the tone of its rccurrence in my
future itfe* The fact that I shall not be able to remember then does not
prcveat ine from anticipating
272 M. C. Sterling

So far äs I can teil from examining what is essential to the occasions

on which I myself am conccrncd or indifferent about future joy or suffer-
ing, it appears that two conditions are necessary and, jointly, sufficient to
justify concern äs opposed to indifference. I am justified in being concerned
about a future experience of joy or suffering just in case: (1) I believe that
such an experience is going to kappen, and (2) that it is I who must ex-
perience it. Given that these two conditions are satisfied in any particular
case, I find it difficult to understand why I would be justified in being
indifferent about the joy or suffering in question.
If believing in eternal recurrence logically compels me to believe, also,
that my present experiences of joy or suffering are going to kappen again
in the next recurrence of my life, and if it implies, further, that it is I who
must experience this joy or suffering, then it follows, on the foregoing
view, that believing in eternal recurrence does in fact have psydiological
consequences such that I am justified in being concerned, and not indif-
ferent, about the predicted experiences. But, is this not precisely what is
implied by eternal recurrence; does it not follow, on the doctrine of eternal
recurrence, that all of my experiences are going to happeri again infinitely
many times, and that it is 7 who must have such experiences? I think the
answer to this question is clearly yes; and although the consequent "human
import" of the doctrine is not necessarily so extreme äs Nietzsche's
"greatest stress," it is certainly a far cry from Soll's "complete indifference."
Soll, doubtless, would agree that eternal recurrence entails that all of
my present experiences are going to happen again in the future; however, in
the passages from his article cited above, there are at least hints that he
would not concede that eternal recurrence allows me to say that it is I who
must have these future experiences. If he is suggesting this, I should like
to argue that he is mistaken, and this is where my next question becomes
relevant: What sorts of considerations might conceivably have led Soll to
believe that if I shall not be able to remember my present life when it
recurs in the future, then its recurrence should be a matter of indifference
tome? ·
Soll's essay seems to presuppose a certain answer to the problem of
personal identity, and I believe it was mainly this presupposition which
led him to the conclusion that one should be indifferent about future
recurrences of one's joy or suffering. For he suggests that there is a sense
of 'identicaP in which it is true to say that I am not identical with that
person that eternal recurrence predicts will "come back" and have the
experiences which I am now having, Moreover, from the fact that I am not
identical with that person, it would have been easy for Soll to infer that
 Recent Discussions of EternsJ Recurrencc 273

I should be indifferent to his suffering or joy. Consider, once more, the

following passage:
Since I am not connected to my recurrcnce in another cycle by being
united with him in one consciousness through links of memory, l am not
identical to bim in the sense necessary for me to view his suffering pain
tantamount to my suffering further pain.10
Soll's reason for maintaining that I am not identical with the future
recurrence of myself is that there is no continuity or accumulation of mem-
ory between myself and "him," Perhaps what is implicitly assumed here
is a quasi-Humean conception of the "seif,* according to which one's
personal identity consists in a collection of experiences whidi are united
through "links of memory*j thus, if there were no links of memory be-
tween the collection of experiences constituting my present life and that
constituting its future recurrence, one could infer that two distinct "per-
sons" or "selves* were involved, and not merely one. But, whatever his
assumpdons about personal identity, Soll does in fact contend that because
I am not united to my future recurrence through links of memory, I am,
in some sense, not identical with him.
Now, if one carefuliy examines the passage originally cited, it will
become apparent that Soll's reason for saying that there can be no
accumulation of memory between cycles is that "one's recurrence in
another cycle is perfectly identical with one's present seif .. " Moreover,
in asserting that strict identity between recurrences implies that there can
be no memory of a past recurrence, Soll is obviously correct; but consider
the logical difficulty whidi this gets him into. For, his conclusion that
there can be no memory of past recurrences rests on his assumption that
recurrences are "perfectly identical"; and, from this conclusion he tries to
establish the negation of prectsely that proposition on whidi the conclusion
rests. More specifically, he first attempts to show that there can be no
continuity of memory between myself and a future recurrence of myself
because the future recurrence of myself and I are perfectly identical, and
then gocs on to ciaim that because there is no such continuity of memory,
I canaot be identical with the future recurrence of myself. Thus, para-
doxically, I both ani and ani not identical with the future recurrence of
myself.
But, maybe Soll's argument cannot be set aside so easily. It is possible
that he vaguely senscd the presetice of this difficulty; for, m claiming that
l am not idenrical with niy future rapurrence, he does a bit of hedging. He
contends that I am not identical with my future recurrence in a spccial

10
Solornon, o/>, */>., p. 340, (Tbc italics are rame).
274 M. C Sterling

sense of the word 'identical/ namely "the sense necessary for me to view
his suff ering pain tantaniount to my suffering further pain."
Unfortunately, Soll does not further specify the sense of 'identicaF
whidi he has in mind; moreover, I seriously doubt that the sense in whidi
I must be identical with my future recurrence in Order to view his suffering
äs equivalent to my suffering is any other than the Standard sense of
'identical/ For, what is necessary for me to view his pain äs equivalent to
my pain if not, simply, that he and I be one, i. e., if not, simply, that it be
I who rnnst experience that f uture pain?
I should like, at this point, to turn to another aspect of SolPs main
thesis, whidi, I believe, militates against its relevance äs an objection to
Nietzsche. In expressing his main thesis, Soll invariably uses the word
'should'; specifically, f uture pleasure and pain in a recurrence of my pres-
ent life should be a matter of complete indifference to me, according to
Soll. Since it is formulated in this fashion, SolPs objection fails to have its
intended applicablility to Nietzsche; for, Nietzsdie's diief point seems to
have been, not so müdi that the thought of eternal recurrence should or
should not weigh upon one's actions "äs the greatest," but rather that it in
fact does weigh upon one's actions ccas the greatest stress."
Next, I should like to make reference to the important, though per-
haps philosophically uninteresting, factor of individual differences. It
seems that the pertinence of this consideration has been overlooked by both
Soll and Nietzsche. I submit that whether or not believing in eternal re-
currence weighs upon one's actions "äs the greatest stress" depends, at least
in part, on who it is that believes it, and what sort of personality he pos-
sesses. Surely ä psychopath, with his low susceptibility to both anxiety and
pain, would be more likely tö. be indifferent about the eternal repetition
of his suffering than would someone with a psychoneurotic bent. Moreover,
I think the majority of people would fall somewhere along the continuum
between these two extremes.
Lastly, the vast remoteness of the future life whidi eternal recurrence
promises me renders the joy and suffering therein ä matter of less concern
than would be the case if this future life were more imminent. Analogously,
if a planet-wide nuclear holocaust were going to occur ten years from
now, I would be concerned about it, but less concerned than I would be, if
it were going to happen ten minutes from now.
In summary, I believe that Soll is wrong in inaintaining that because
of the "suprahistorical character" of eternal recurrence, one should be
indifferent about future joy or suffering in a recurrence of one's present
life; however, even if this were true^ it is qüestionable whether it has any
bearing on Nietzsche, who is principally making a factual claim. Further-
 Recent Discussions of Eternal Recurrence 275

more, contrary to both Soll and Nietzsche, I think that the factor of
"individual differences* has some influence on the extent to which be-
lieving in eternal recurrence weighs upon one's actions "äs the greatest
stress," And, lastly, I believe that the sheer remoteness of the future re-
currence of one's present life diminishes, at least to some degree, the impact
which believing in eternal recurrence would have.

III
In this section, I shall summarize the main contentions of an essay
by Arnold Zuboff11 entitled Nietzstäe and Eternal Recurrence. The general
question which Zuboff deals with in this essay, though he does not explicit-
ly formulate it, is: Does the doctrine of eternal recurrence have personal
significance such äs would justify a Nietzsdiean revaluation of all values?
His conclusion, broadly speakLng, is that the doctrine of eternal recurrence
is lacking in such human Import and, consequently, that Nietzsche's
assertions to this effect are false.
Before getting into the essay itself, I should like to digress, for a mo-
ment, to clarif y the expression 'revaluation of all values/ 'Revaluation of all
values' is the usual, English translation of Nietzsche's expression 'Umwer-
tung aller Werte/ The prefix cum* in German often signifies a "reversal" of
some sort; and *to revaluate/ I think, means the same äs ffto diange one's
opinion about the worth, importance, or value of something'; moreover,
a specifically Nietzsdiean revaluation is a change of one*s opinion about
the importance, worth, or value of this life, with its actions and experiences,
äs opposed to life in a "beyond" of some sort. It implies affirmatively
embracing thh life; dierishing and living it to the füllest; and regarding its
dccisions and actions äs having the utmost importance* The Nietzsdiean
revaluation involves a repudiation of traditional, Platonic-Christian other-
worldliness, and the cxaltation of what we might (somewhat barbaricly)
call an Anthropocentric-Atheistic lA/sworldliness.
Let us now turn to Zuboff's essay itself. Towards the beginning of his
essay, Zuboff calls anention to a difficulty with eternal recurrence of
whidi he thinks Nietzsdie may not have been aware. Specifically, he sug-
gests that there are at least two non~Nietz$diean interpretations of eternal
recurrence, which negate the personal significancc it must have in order to
justify a revaluation of all values. For rcasons that will be discussed below,
he designatcs thesc t^^o aon-Nietzschean interpretations of eternal recur-
rence o$ the Äin$uIatingÄ Version and the ^Leibnizian* vemon* After
describing thesc and attcmpting to show why they thwart Nietzsdic's
st
Solomon, Roben: C^ op. cit», p. 343*^357.
276 M. C Sterling

revaluation projcct, Zuboff turns hb attention to Nietzsche's own version.

Hc argues that, on this version äs well, eternal recurrence is deprived of the
personal significance that it needs in order to justify a Nietzschean reva-
luation. I shall consider, in turn, what Zuboff has to say about these three
versions of eternal recurrence.
Let us first consider the "insulating" version. On this version of eternai
recurrence, the "recurrence of a life means not the repetition and return of
the same man, but rather the generation of a series of men, eadi a mere
duplicate of the last." What is distinctive about this version is that, if it
is true, then Zuboff, for example, is in no sense numerically identical with
any of the previous or future Zuboffs entailed by eternal recurrence, and
these latter are all numerically different fröm one another. Thus, what is
involved here is not one person at all, but rather an infinite number of
completely distinct, though exactly similar persons; to use Zuboffs ex-
pression, the recurring individuals wöuld be numerically "insulated" from
one another. .^
As for the Leibnizian version, we might think of that äs being the
extreme opposite of the insulating Interpretation. It asserts that the ex-
periences of this present life are absolutely identical with their alleged
"future" recurrences and, consequently, that one cannot consistently speak
of there being any experiences over and abpve, i. e., in addition to the
experiences of this present life, This, Zuboff teils us, is an application of
Leibniz's Law. Leibniz's Law says that if an entity, x, has all properties
in common with an entity, y, then it is pointless to talk about t wo entities;
for and y, being indistinguishable, would be numerically identical, i. e.,
one. On the Leibnizian version, the experiences of this present life simply
are the experiences of that future life entailed by eternal recurrence.
Now, what Zuboff is especially anxious to emphasize about these
non-Nietzsdiean versions of eternal recurrence are their alleged devas-
tating consequences for Nietzsches "revaluation project." He thinks that,
on either of these interpretations of recurrence, its personal significance
is so greatly diminished that it can no longer effect a revaluation of all
values. (Perhaps I should interject here a few words about w!hat, I take it,
is meant by the expression 'personal significance/ So far äs I can teil, to
say that the doctrine of eternal recurrence has "personal significance" is to
say, broadly speaking, that it entails certain propositions about human
experiences [especially, experiences of joy or süffering]. To speak more
specifically, it is to claini that each human being may truthfully say the
following: The doctrine of eternal recurrence entails an eternalization of
my experiences. The insulating Interpretation eliminates "personal signifi-
cance" because, here, eternal recurrence no longer entails the eternalization
 Reccnt Discussions of Etcrnal Rccurrcncc 277

of my experiences. And the Leibnizian Interpretation does away with

"personal significance* because, in this case, eternal recurrence does not
entail, in the requisite sense, an eternalization of my experiences.)
The Leibnizian shares with the insulating treatment a rejection of the
personal significance that Nietzsche gives recurrence» For the insulating
theory the recurrences bring only otber experiencing beings; for the
Leibnizian they bring nothtng additional to a man's present experience
— in fact, the recurrence$ already are in present experience — present
experience already is in the recurrences. If a man cannot look forward
to his recurrence äs being for him an additional experience either because
the experience within it will not be his or because it is his now, the
recurrence cannot hold the awesome promise or threat that gives it
reforming power.12
The Nietzschean Interpretation of eternal recurrence, however, avoids
the drawbacks of both these non-Nietzsdhean interpretations. On the
Nietzschean Interpretation, it is / who will come back, and re-experience
what I now experience; this preserves the personal significance of eternai
recurrence from the destructive consequences of the "insulating treatment/*
Furthermore, the experiences I shall have upon coming bade are to count äs
other tban, i. e., äs in additlon to what I now experience; this defeats the
Leibnizian Interpretation and, again, eternal recurrence retains its personal
significance.
Nonetheless, Zuboff believes there to be other difficulties in Nietz-
sche's Suggestion that eternal recurrence makes possible a revaluation of all
values. To show this, he first considers the consequences for personal
significance of rejecting determinism, and, then, the consequences of
retaining it. He argues that in both cases personal significance is nullified
by the possibility of "many variations of a man's life along with the many
recurrences.* In the remainder of this section, I should like to take a closer
look at what Zuboff is arguing here; get clear about it; and state any
criticisms that seem appropriate,
Let us first consider ZubofPs Suggestion that, äs a consequence of
rejecting determinisni, there would be many * variations of a man's Üfe"
äs well äs recunrences, What does this mean? What is a "Variation of a
man's life?* What is a recurrence of a man's life?
It seems fairly certam that Zuboff, in making the above claim, is
thinking in terms of Nietzsae's ontology of power-centers. In view ojf
ihis ontology, then, how do we answer the abovc questions?
Let us suppose, äs Nietzsche did, that the universc consists of a finitc
number of power-centers, and that all ^things," consequently, are ekher
sample powcr-ccnters or complexes of intüracting power-centers. Now, in
« Solornon. op. dt., p. J5i i(TEc Julies are miac).
278 M. C. Störung

a universe such äs this, what are sorae of the ways in which we can think
of personal identity? Just offhand, I can think of at least three ways in
which personal identity might be thought of.
Firstly, we might, a la Berkeley, think of the man himself, i. e., the
"person" or 'T' itself, äs being identicäl with a single unit of what there is
— in this case, a single power-center. Alternatively, we might, a la Greek
atomists, identify the man himself with a particular assemblage of what
there is, i. e., power-centers, and, lastly, we might, ä la crude materialists,
regard the " * äs being a by-product of a particular pattern-of-activity of
what there is, namely power-centers.
If we suppose the f irst alternative to be true, then Zuboff, f or example,
is not strictly identicäl with the entire set of power-centers of whidi he, äs
a complex organism, is compounded; rather, he is identicäl with oftly orte
among this enormous number of power-centers. Let us call this one power-
center with which Zuboff himself is identicäl the Zentral power-center/
Now, although, at the moment, this power-center "occupies" the central
Position in one particular organism (i. e., complex of power-centers), it is
conceivable that i t might become separated from this organism; moreover,
we might imagine its becoming the central power-center of any of a large
number of other possible organisms. Furthermore, it is conceivable that the
experiences of this power-center during each "occupancy" (in these various
possible complexes of power-centers) might be altogether dif ferent. On the
other hand, it is also conceivable that this pöwer-ceiiter, after separating,
might once again become "central* in the same complex of power-centers
with which it is now in interaction; we might even imagine that, in some of
these cases of "ire-association," this power-center might experience the same
sequence of experiences that it had during a previous instance of inter-
action.
On the above assumptions, Variation of a man's life' could be under-
stood äs meaning any case of central "occupancy" in a complex of
power-centers other than the complex of present "occupancy." As for
c
recurrence of a man's life/ this might be taken to mean a future case of
central "occüpancy" in this complex of present "occupancy," in which-die
central power-center again has the same sequence of experiences äs now.
Something similar might apply, on the second of the above alter-
natives, i. e., that in whidi the *I* is identicäl with a particular assemblage
of power-centers. Here, we could think of the. assemblage äs a distinct
System capable of entering into association with many different complexes
of power-centers (i. e,, organisms). A Variation of this life, then, would be
its entrance into association with power-complexes other than the complex
of its present association; a recurrence of this life would be its re-entering
 Recent Discussions of Eternal Recurrence 279

into association with this present complex of power-centers, where the

same sequence of experiences is repeated in the central power-center.
On the third alternative, if any group of power-centers were to come
together and to behave in a certain manner, their interaction would gener-
ate or give rise to the particular al* now known äs Arnold Zuboff. When-
ever and wherever this transient "I" existed it might become associated
with many complexes of power-centers other than that with which it is now
associated. Association with complexes other than the present complex
might, in this case, be the ineaning of Variations of a man's life/ and re-
association (with an identical sequence of experiences for the central
power-center) with this present complex might be the meaning of 'recur-
rence of a man's life."
Now, let us return to Zuboff's Suggestion that, in a non-deterministic
universe, eternal recurrence would be consistent with the occurrence of
many variations of a man's life äs well äs recurrences of it*
In a non-deterministic universe, there would be nothing to guarantee
that a particular state-of-affairs or event must invariably be followed by
the same sequence of states-of-affairs or events. Thus, on any of the above
conceptions of personal identity, this would mean that, in a non-deter-
ministic universe, the same al" or "person* might on countless future
occasions "corne back* into association with the complex of power-centers
with which it is now associated, and yet have a sequence of experiences
different from that had during any previous period of association with
this complex,
For exarnple, the alw (i.e., the simple power-center; assemblage of
power-centers; or by~product of power-center activity) currently known
äs Zuboff may at some point in the future re-enter into association with the
particular group of ppwer-centers with which it is now associated. Yet, if
this were to happen, (assuming there is no determinism), the rest of the
universe might, at that tirae, be in a completely different condition than
it is now. This re-integration of the organism, moreover, might take place
on some planet other than Earth; Zuboff might enter some other profession
besides academic philosophy; in a word, his life might run a completely
different course from what it is now running.
Now, if by ^Variation of a man's life' we also understand such cases
of rc-integration äs that just describedf then it seems that eternal re-
currence would, given a non-deterministic universe, be consistent with
there being many variations of a man's life äs well äs recurrences of it*
What Zuboff specifically wams to $ay is that, in such a universe,
diese countless Variation* of a man's life would so grcatly outweigh the
occasionai and higjhly improbable r«rurraice of it that he need have no
280 M* C Sterling

personal cöncern about the possibility of being eternally rewarded or

punished äs a f unction of how he Kves this life.
In a non-deterministic universe, eternal recurrence would mean an
eternal coming-back of the same states of the universe, but not their eter-
nally coming-back in the same order, Thus, a particular life might actually
occur once and never return, if a specific permutation of experiences is at
least part of what constitutes the sameness of a particular life. Furthermore,
since the "reforming power" of eternal recurrence depends on the eternal
coming-back of precisely this life, I think that Zuboff is correct in asserting
that Nietzsche's revaluation project would be "wrecked," in such a universe
äs this.
Zuboff, however, is not content to /maintain simply that eternal
recurrence is deprived of its reforming power in a non-deterministic
universe. He goes on to assert, äs well, that eternal recurrence does not
justify a Nietzschean revaluation even if determinism is the case; and
here is where I think he is mistaken. He reasons äs follows:
This concept, whidi I find fascinating in itself, wrecks Nietzsdnfs me of
recurrence in bis revaluation project once variations äs well äs repeti-
tions of eadrman's life are allowed to be woven into the fabric of the
universe....
But, I am afraid, even the exclusiveness of the Nietzschean ring is prob-
ably not enough to save his concept froni the fate I have indicated. For
even if we accept his [Nietzsche's] deterministic circle äs the sole reality,
I would expect many variations of a life.to pop up in such a world
before it gets around to repetitions anyway. Suppose we think in terms
of Nietzsche's own belief in cycles of destructions and re-creations of
the world. It seems unlikely to me that the re-creation of the world
after the next destruction would result in a precise repetition öf the
world äs it is now, even if there are only finite possible arrangements
of forces in the universe; It seems to me that there would probably be
vast numbers of destructions and re-creations before a repetition was
struck. But by then, we might expect any number of variations of a life
to have been locked into the ring along with the repetition. After all,
the odds against any versiofi of the world being precisely lik<e this one
are immensely higher than those against its simply containing one of
countless possible variations of a life.18
Through the variations alone my life is continued and extended.
This unimaginable expansion of my life swamps the significance of any
repetitions that happen to come up of the life I live here and now. And
so Nietzsche has lost.14
Now, it seems to me that even if determinism is consistent with
there being many more variations of this preseiit life than exact recurrences

13
Solomon, op. cit., p. 351. (The italics are mine).
14
ibid., p. 357 (my italics).
 Receiu Discusslons of Eternal Recurrence 281

of it, it does not follow, äs Zuboff appears to believe, that eternal re-
currence would be powerless to ef fect a revaluation of all values. For, even
though tfaere were many more variations of this present life than exact
reperitions of it, the$c variations themselves would, in a deterministic
universe, be exactly repeated eternally.
For instance, between now and the next, exact recurrence of my
present life I might live a tremendous number of variations of it. None-
theless, in a deterministic universe, eternal recurrence would imply that all
of the interim variations of my present life will also exactly recur, in
precisely the same order, eternally, Thus, Nietzsdie's injunction to "So live
that you must desire to live again" would be applicable, in the same de-
gree, to this present life (for it will recur eternally); to all variations of
this present life (for they will recur eternally); and to all the exact
recurrences of both this life and its variations (for they will recur eternally).
Moreover, obedience to this, Nietzsdie's Eternalistic Imperative, is the
ultimate behavioral outcome of a Nietzschean revaluation of all values.
In conclusion, I think that Zuboff is right in claiming that, in a
non-detenninistic universe, eternal recurrence would not justify a Nietz-
schean revaluation; however, his argument that the same would be true in
a deterministic universe strikes me äs unconvincing.

IV
Bernd Magnus has recently expressed some interesting views on
Nietzsdie's doctrine of eternal recurrence.18 His writings on the subject
are relatively extensive, covering many different facets of the doctrine;
consequently, I have restricted my discussion to a few selected ideas whidi
strike me äs particularly interesting. In this section, I shall first briefly
dclineatc these selected ideas; then, I shall try to examine them critically.
To begin with, I think the views expressed in the following passage
are noteworthy:
NIetzsdie writes in his notes: eMy doctrine declares: the task is to live
in such way that you must wish to live agam — you will anyway."
When die normative vcrsion of eternal recurrence is interpreted in a
literal fashion analogous to the Interpretation of the cosmological
version, severe internal inconsistencies arise. How is the regulative read-
ing of the doctrine of eternal recurrence to be rcconciled with an
eternally recurring and objectivc identical life? If our lives are in fact
bist eternal repetitions» containing the same anguish, cxultation and

:5
Magnus, Bernd, *Nitui<he'* Etcrnalistic Coonter-Myth* Rcview &} Mctapky*ic*> Vol.
XXVI» JUGC 1973, p$> 604—616, Also: Nmzttbc'i Ettrnalittic Count*r~Myth (un-
äs of thi*
282 M. C. Sterling

monotony, does it make sense to postulate an imperative enjoining us to

live äs if our lives recur eternally? If our lives äs now experienced are
but repetitions of an infinity of lives identically endured, is there not a
determinism implied in this doctrine, thus vitiating all imperatives? If
eternal recurrence expresses a "naturai kw," how can it become an
object of choice? It would, on the surface, seem äs meaningless to insist
that we live "äs if" there were an eternal recurrence äs to admonish us
to live "äs if * subjected to the law of gravity.
The assertion that life recurs eternally (empirically) is difficult to
reconcile with the exhortation to live äs if it recurred eternally.16
To get clear about what Magnus is saying in this quotation, and to
make explicit the sort of reasoning involved here, let us first consider the
two versions of Nietzsche's doctrine which this passage presupposes.
Magnus, along with many other Nietzsche-sdiolars, makes a distinction
between the doctrine of eternal recurrence äs a cosmology, and the doctrine
of eternal recurrence äs an ethical imperative. As a cosmological thesis,
eternal recurrence purports to assert a fact about the universe; specifically,
it asserts that whatever occurs in the universe (hence, whatever and indi-
vidual experiences) has occurred and will recuf infinitely many times. As
an ethical imperative, Nietzsche's doctrine constitutes a (im)moral com-
mand exhorting us to live in such a way that we must desire to live again.
In the above passage, then, Magnus is basically making a claim about the
compatibility of these two versions of eternal recurrence.
Broadly speaking', he is claiming that if all our experiences in fact
recur eternally, then it does not "make sense" to issue an imperative
commanding us to live äs if all our experiences recurred eternally. Magnus
believes that enjoining us to live äs if our lives recurred eternally, when
in fact they do recur eternally, is äs "meaningless" äs commanding us to
live äs if subjected to the law of gravity, when in fact we are obviously
subject to that law.
On the surface, this might seem an odd sort of claim to make; if,
however, one takes a more careful look at the cited passage, what Magnus
is saying becomes a bit less perplexing. His point seems to be that it is the
determinism entailed by Nietzsche's teaching (construed äs a cosmology)
that renders it meaningless to issue an imperative enjoining us to live äs if
there were an eternal recurrence. His reasoning, if I am not mistaken,
would be roughly äs follows: Only if human beings are free does it amake
sense * to issue them imperatives; the doctrine of eternal recurrence, äs a
cosmology, implies determinism; determinism, in turn, implies that human
beings are not free; consequently, if eternal recurrence, äs a cosmology, is
true, then it does not make sense to issue imperatives to human beings. In a
16
Magnus, Bernd, op. cit.> p. 608.
 Recent Discussions of Eternal Recurrencc 283

word, the doctrine of eternal recurrence, äs a cosmology, implies a "deter-

minism thus vitiating all imperatives." So, to understand Magnus* claim,
what it is crucial to note, I think, is: (a) the fact that eternal recurrence,
construed äs a cosmology, implies determinism, and (b) Magnus' apparent
presupposition that human freedom is incompatible with determinism.
The next passage that I shall consider is the following:
Apart from the internal dynamics of Nietzsdie's own thinking, what
licenses the doctrine of eternal recurrence? Nietzsdie claims that if we
genuinely appropriate, i. e., internalize the doctrine of eternal recurrence,
it would lay the greatest stress upon our actions. But is that true? I can
think of at least two objections to Nietzsdie's claim.
If we adopt the Absurdist frame of reference and accept the inescapa-
bility and finality of death, we might formulate the following question
to ourselves: Do you want this once and once only?> rather than "Do
you want this once more and innumerable times more?" Surely it could
be argued that internalizing tbis question will also weigh upon our
actions äs the greatest stress. If I genuinely believe that this life is all
there is, it might help to sort out the trivial from the important, what
sustains me from what merely distracts me. If there is any wisdom at
all in the youthfui exhortation to edo your own thing,* it lies, it seems
to me, in the challenge to find what is genuinely yours, your own, and
to reniain faithful to that. It is a twendeth Century reincarnation of the
Delphic injunction: Know thyself! Now, discovering what is truly yours
can be equally painful and difficult whether the discovery enriches one
liferimc or infinite repetitions, at least so it seems to me. There is, in
addition, an obvious reason for accepting the hypothcsis of radical
finicudc instead of eternal recurrence, and that is diät one is true and
the other is fiction.17
In this passage, Magnus seems to be expressing the view that it is not
true, äs Nietzsche believes, that "internalizing" the eternal recurrence
notion would lead to its laying the greatest stress upon our actions. To show
this, he calls our attention to the consequences that would follow upon one's
appropriating the idea that death is inescapable and absolutely final. If we
were to accept this thesis of "radical finitude,* he suggests, surely we
should be deeply affected and should, perhaps, lead our lives a bit dif-
fcrently than we should have, if this idea had not impressed itself upon
us. Moreover, according to Magnus, the *thesis of radical finitude* has the
dedded adirantage over Nietzsdie*s idea of being true; the latter idea, he
says, Äis fiction.1*
Some vcry interesting questions are raised by this passage; I shall
return to them shortly.
Now, let us take a look at another passage in whidb Magnus toudies
upon a very signif icant issuc:
of. cii
234 M. C Sterling

There is another and perhaps strongcr reason for rejecting the eternal
rccurrence hypothesis. The yoke of ctcrnal recurrence is shattering,
according to Nietzsche, because our actions are condemned to infinite
repetitions, But it i$ difficult to see how I can identify with that person
who performed my actions an infinity of times before and will presum-
ably perform them an infinite number of times hereafter. Perhaps an
analogy will help. Let us assume that somewhere in the universe there is
a planet just like this one. At this moment there is a person on that
planet named Bernd Magnus, who is writing this paper at UGR, for an
audience exactly like this one. One of the things he would be thinking
right now is that he is not me, even though an identity of indiscernibles
obtains between us. And I think he would be right. I think he is right
because we are at least numerically (and perhaps spatially) distinct.
Now how does this apply to Nietzsche*«' theory? Even though recurren-
ces are not simultaneous, the fact remains that we recur at different
points and times in the cosmic series. But recurring at different times
within a series is sufficient, in my opinion, to defeat the identity. That
is what I meant when I said earlier that if someone who in all other
respects is exactly like me had written this sentence during a previous
recurrence, I should disown him.18
Here Magnus appears to be arguing that because of certain con-
siderations about personal indentity, the psychological impact of eternal
recurrence, and hence, its power to influence our conduct, should be
negligible. More specifically, he suggests that since a man is not identical
with his future recurrence, eternal recurrence should not motivate him to
refrain from performing those actions whidi he could not will eternally;
it should not compel him, that is, to live in such a way that he must desire
to live again.
The reasoning which Magnus offers in support of this view is thought-
provoking; he asserts that occurring at different times in a series is suff icient
to defeat identity between oneself and the future recurrence of oneself.
Moreover, if one is not identical with one's future recurrence, then (bar-
ring an uncommonly compassionate nature) one should be indifferent to
his experiences. Contrary to Nietzsche, then, Magnus does not believe that
eternal recurrence should weigh upoii our actions äs the greatest stress.
Having outlined Magnus' views on several, important aspects of
the eternal recurrence notion, I should like now to consider them more
closely. In the upcoming discussion, I shall refer to the first quotation äs
passage A; to the second äs passage B; and to the third äs passäge C.
As indicated, the main idea expressed in passage A is that the cos-
mological and ethical-imperative versions of eternal recurrence are in-
compatible. Let us first focus attention on the Statement: "It would, on the

18
Magnus, op. cit., 614—615.
 Reeent Discussions of Eternal Recurrence 285

surface, seem äs meaningless to insist that we live 'äs iP there were an

eternai recurrence äs to admonish us to live *as if subjected to the law of
gravity." Now, this Statement suggests that it would be meaningless to
exhort us to live äs if subjected to the law of gravity, and, moreover, that
it would be equally meaningless to exhort us to live äs if there were an
eternal recurrence. What Magnus means, I think, is that just äs we bave
no choice but to live äs if subjected to the law of gravity, so also (if eternal
recurrence is true) we have no choice but to live äs if eternal recurrence were
true. This certainly sounds plausible enough; but is it true?
It seems to me that there is a sense in which Magnus is perfectly
correct. Cleärly, it would be pointless and unnecessary for someone to
command us to live äs if subjected to the law of gravity; for surely there
is a sense in which we simply mu$t live äs if subjected to this law. More-
over, there is a sense in which the same consideration would apply to some-
one's insisting that we live äs if there were an eternal recurrence (if, in fact,
eternal recurrence were true)· Nonetheless, I think that insofar äs Magnus'
Statements are intended to illustrate the incompatibility of Nietzsche's
cosmological and ethical pronouncements, they are misguided.
If I am not mistaken, Magnus* diief difficulty here is that of equi-
vocation. For, I believe I detect that two different senses are being con-
flated, both in the case of the expression *to live äs if subjected to the law
of gravity/ and in that of the expression *to live äs if there were an eternal
recurrence/ Let us try to "disambiguate" these expressions.
Firstly, we need to ask: What sorts of things might count äs instances
of living äs if subjected to the law of gravity? I submit that there are at
least two distinct sorts of things that might count äs instances of living
äs if subjected to the law of gravity. On the one hand, a person might be
said to be living äs if subjected to the law of gravity just in case his body
actually behaved (or was disposed to behave) in the wäy Newton predicted.
Accordingly, if this person, upon being pushed from the top of UCR's
bell-tower, proceeded to move toward the earth in the appropriate manner,
we might say that his behavior counted a$ an instance of living äs if sub-
jected to the law öf gravity- Moreover, in this sense of the expression
'living a$ if subjected to die Jaw of gravity/ one could legitimately claim
thai it would be pointless to command us to live äs if subjected to the law
of gravity. For we hav£ no dicke but to live äs if subjected to the law of
gravity, in this sense of the expression. Up to this point, then, Magnus is
right.
On the other hand, a person might be said to be living äs if subjected
to the law of gravity just in case he u$ed his fanruliarity with this law to
nchicvc tltose rcsulrs hc desircd and 10 avaid thosc which he found un-
286 M. C. Sterling

desirable. For example, such a person might avoid jumping off UCR's
bell-tower; or hc might tightly secure the belt of bis over-sized trousers; or
he might, upon noting that the cord connecting a heavy chandelier to the
ceiling was about to succumb, suggest to his friend, who is Standing be-
neath the chandelier, that he make haste to relocate himself.
All three of these cases, I believe, might count äs instances of living
äs if subjected to the law of gravity. Furthermore, in this sense of the
expression fliving äs if subjected to the law of gravity/ I think it would
make perfectly good sense to exhort one to live äs subjected to the law of
gravity. So, in this sense at least, I suspect Magnus is wrong in suggesting
such an exhortation to be meaningless.
Now, what about the expression 'livirig äs if there were an eternal
recurrence?' What sorts of things count äs instances of living äs if there
were an eternal recurrence? Again, I think there are at least two distinct
senses in which one might be said to live äs if there were an eternal re-
currence. On the one hand, one might be said to be living äs if there were
an eternal recurrence just in case his life kept cropping up again and again,
i. e., just in case his life, in fact, were to keep "coming back" eternally. On
this analysis, if eternal recurrence were true, it would be pointless to
command us to live äs if it were true; for that would only amount to
commanding us to do what we are bound to do anyway. In this sense,
Magnus' claim would be true.
On the other hand, the expression cliving äs if there were an eternal
recurrence' might mean something quite different. In this other sense of
the expression, a person would be said to be living äs if there were an
eternal recurrence just in case he: (1) put forth the greatest possible effort
to maximize his joy in this life, and (2) put forth the greatest possible
effort to minimize his suffering in this life. On this Interpretation, even
though there were in fact an eternal recurrence, it would still make sense,
I believe, to issue an imperative enjoining us to live äs if there were an
eternal recurrence. Indeed, the fact of there truly being an eternal
recurrence would be precisely what validated the issuance of the im-
perative, and justified our obeying it. In this sense of 'liviiig äs if there
were an eternal recurrence/ what better reason could there be for the
making and the heeding of such a command than the fact that there
actually was an eternal recurrence? If, in fact, all of may expefiences will
be repeated eternally (and if my voluntary behavior is, at least in some
degree, instrumental in determining the diaracter of my experiences), then
would I not be wise to put forth every effort to maximize my joy and
minimize my suffering in this life* i. e., would I not be wise to live äs if
there were an eternal recurrence?
 Recent Discussions of Eternal Recurrence 287

The Nietzsdxean ethical imperative commands us to live our lives äs

if there were an eternal recurrence; moreover, it seems to me that what he
meant by this command is in no way inconsistent with his simultaneously
advancing eternal recurrence äs a cosmology. In enjoining us to live in
such a way that we must desire to live again, I think Nietzsche is simply
saying that since we shall in fact live again, and since the joy and suffering
of our future lives will be commensurate with the joy and suf fering of this
life, we would be prudent to put forth every effort to maximize our joy and
to minimize our suf fering in this life.
Thus, I believe that Magnus is wrong in regarding the cosmological
version of eternal recurrence äs incompatible with the ethical-imperative
version. His basic problem, I think, stems from confusing two distinct
senses of the expression 'living äs if there were an eternal recurrence.*
But there still remains the difficulty posed by Nietzsche's determinism.
For it might be argued (äs Magnus appears to be arguing) that the cosmolo-
gical version of Nietzsche's doctrine implies a determinism whidh, in turn,
implies that there is no freedom; and if there is no freedom, then we can-
not freely dioose to obey imperatives; moreover, if we cannot freely
choose to obey imperatives, then it would seem pointless for someone to
issue an imperative. On these assumptions, we would have to continue
regarding the cosmological and ethical-imperative versions of eternal
recurrence äs incompatible; for if the former version were true, the latter
would be pointless. Eternal recurrence äs a cosmology would, in Magnus'
words, imply a determinism, *thus vitiating all imperatives.*
It should be noted, however, that the analysis of 'freedom' presup-
posed by this argument is not the only one conceivable. The line of
reasoning sketdhed above presupposes an analysis of 'freedom' in terms of
indeterminism; in other words, on this analysis, 'free act* means the same
thing a$ rfact arising from no cause whatsoever/ The transition from the
assurnption of determinism to that of the non-existence of freedom depends
on this anaiysis.
However, I think the mdeterministic analysis of *freedom* has the
disadvantage of being inconsistent with ordinary usage; for it seems to me
that when we ay, for instance, that Jones freely did x, what we are claim-
ing, among other things, just is that i$ determined. When we say that
Jones* action, x, was free, part of what we are claiming is that it was Jones
that did x, L e., it was Jones that caused to happen; and to say that
Jones caused to happen, it seems to me, amounts to admitting that had
a cause and, hencc, was dctcrmincd.
Thi$ sort of reasoning hints at the appropriatencss of a "compati-
bilist" analysis of 4freedom/ TTie most obvious course for a compatibilist
288 M. C Sterling

to follow would be to define 'free act' äs meaning the same äs cact whidi
does not result from external influence.' On this analysis of 'free act/ it is
not self-comradictory to say that a given act 1s botb free and determined.
Moreover, if we unpack what is still implicit in this analysis, namely that
a free act is an act resulting from something (a cause?) other than external
influence, then we may claim, not only that the same act can be consistent-
ly called both 'free' and 'determined/ but also that a free act is by definition
a determined act.
My aim here is not to offer a simplistic solution to so complex a
thing äs the freedom-determinism issue. My intention is rather to point
out that the legitimacy of Magnus' apparent belief that determinism
vitiates all imperatives rests oii a specific analysis of 'f reedom/ and that
since other analyses are possible, Magnus' assertions about the consequent
incompatibility of Nietzsche's cosmological and ethical Statements are
legitimate only if "his" analysis happens to be the right one.
Let us, now, take a look at passage B. The first thing that Magnus
claims in this passage is that if we were to adopt what he calls the
"Absurdist frame-of-reference," this would affect us in the same way in
which Nietzsche alleges that belief in eternal recurrence would affect us.
More specifically, he suggests that äs a result of adopting such a stand-
point, the question "Do you want this once- and only once?" would lay
the greatest stress on otur actions. Here, I think Magnus is calling attention
to a very significant point.
However, in the paragraph just previous to this, Magnus had stated
that he could think of at least two objections to Nietzsche's claim that
internalizing the eternal recurrence notion would lay the greatest stress
on our actions. Thus, Magnus initially gave me the feeling he was about
to show wby Nietzsche was mistaken in asserting eternal recurrence to
have such great impact. But, do Magnus' Statements about the "Absurdist
frame-of-reference" really constitüte an objection to Nietzsdie's claim that
eternal recurrence lay s the greatest stress on our actions? To show that
Nietzsche is wrong, it does not suffice simply to establish that something
other than belief in eternal recurrence has the same effect on us that eternal
recurrence is supposed to have. For ixo contradiction is involved in main-
taining that both belife in eternal recurrence and adoption of the "Absur-
dist frame-of-referencew would weigh upon our actions äs the greatest
stress (that is, unless one quibbles about the Superlative suffix in 'greatest').
I shall not belabor this point, for, at most, it involves a simple logical
mistake of negligible importance. Besides, it may not even have been Mag-
nus' intention to disprove (by allusion to the "Absurdist frame-of-ref-
erence") Nietzsche's claim about the psychological import of eternal re-
 Rcccnt Discussions of Etcrnal Recurrence 289

currence. Let it suffice to säy that, since the opening Statements of

passage B at least give the Impression that Nietzsche is about to be refuted,
Magnus* language, here, is misleading.
One thing that passage B does establish, however, is that belief in
eternal recurrence is not the only means by which a Nietzsdbean "revalua-
tion of all values* might be effected. In other words, B inakes it clear that
Platonic-Christian contempt and disesteem for the things of this life might
be uprooted without appeal to an eternalization of this life. Human beings
might dispense with the idea of a "beyond" which is superior to this life;
more perfect and better than this life; and might come to value, esteem,
prize and praise this life without ever needing recourse to the idea that
this life recurs eternaily and is, hence, in some sense permanent.
In passage B, Magnus cogently illustrates that this life can be assigned
Superlative importance, even when it is assumed to be characterized by the
most radical kind of impermanence. There is the implication, I think, that
the actions and experiences of this life might be regarded äs all the more
important precisely because of a realization of life's finitude, i.e., the
absolute finality of death.
This is extremely interesting; it makes one wonder why Nietzsdie
paid sudi exclusive attention to the cosmological Version of eternal re-
currence in attempting to effect a "revaluation of all values" (thus,
justifying his ethical imperative).
Magnus* question (L e., Why should not life be regarded äs all the
more precious and valuable because of its transience?) would have been a
very tough one for Nietzsdbe to deal with, considering his great attadiment
to the eternal recurrence idea. Moreover, there is an additional and related
problem for Nietzsdie. For just äs an awareness of life*s transience might
eahance its value for us, so also might a realization of its extreme rareness
in the universe. Even if there are countless planets in the universe that are
inhabited by intelligent forms of life, it still seems fairly certain that an
overwhelming proportion of the matter in the universe exists in an in-
organic, lifeless condition. Since the amount of non-iiving matter in the
universe vastly cxcceds that of living matter, it would appear that the
odds against being alive are tremendous, This might tend to make one feel
octrexnely fortunate to be alive, and all the more determined to get the
most out of Hfe while it lasts.
Had Nietzsdie scriously reflected on sudi facts äs thcse, he might have
endeavored to carry out his *reva!uation projcct" in a slightly differcnt
mannen Perhaps he would have concluded that what was needed to bring
about a "revaluation of all valucs* (and, thus, a revcrsal of the Platonic-
Christian tradmon against whidb he so vehemently struggled) was, not an
290 M, G Sterling

"eternalistic counter-myth," but rather a head-on confrontation with the

facts of human existence, and a realization of life's evanescence and
extreme rarity within a largely non-living universe. For, this, he might have
concluded, would lead us to have a greater appreciation for this life; dis-
pose us to live life to the füllest; and to esteem its every moment äs being
utterly precious. In a word, it might motivate us to live in such a way we
must desire to live again, even though we had no illusion whatever that we
actually would live again. We could, thus, effect a Nietzschean evaluation,
and retain the ethical imperative which such a revaluation calls for, with^-
out ever needihg to invoke eternal recurrence äs a cosmology.
As indicated, passage B raises no genuine objection to Nietzsche's
claim that belief in eternal recurrence would weigh upon one's actions äs
the greatest stress. In passage C, however, Magnus points out a potentially
quite serious objection to Nietzsche's claim. He argues that one is not
identical with one's future recurrence. For example, in my own case, this
would mean that I am not identical with that person who, in the next
cycle of recurrences, will have the experiences which I am now having. If
this is true, then the doctrine of eternal recurrence is completely bereit
of the personal significance it would need in order to effect the "revalua-
tion of all values" which Nietzsche desired; instead of weighing upon our
actions äs the greatest stress, then, it would be somethiiig about which we
might justifiably be indifferent.
What is interesting, though, is Magnus' reason for holding this view.
He suggests that I, for instance, am not the same person äs my future
recurrence because he (my future recurrence) will occur at a different time
tban L
Now, what makes it difficult to respond to this (whether by way of
agreeing, or disagreeing) is that Magnus is obviously leaving some impor-
tant assumptions unstated; the most crucial of these, I think, are assump-
tions about personal identity. What is it that ultimately constitutes the
"!/' i. e., the person himself ? Is the *I* an event and, äs such, a fleetirig
sort of entity? Is it, thus, something whidi cän occur and then cease to
occur? Would it be altogether preposterous to regard the " * äs substance,
and, hence, äs persisting throughout all change?
Let us suppose that the el* is an event of some kind. Let us assume,
moreover, that is such an event, and that y is a repetition of k. Now the
fact that y is a repetition of means that y is in some way different from
x; otherwise, there would be no basis for the differing descriptions. But if
that is the ease, can still be said to be identical with y ?
Appareütly Magnus' answer to this question would be: no. I think
the underlying assumption here is that in order for to be identical with
 Recent Discussions pf Eternal Recurrence 291

y, it must be possible to predicate of χ everything that can be predicated

of y; but, clearly, even if χ and y are indiscernible in all other respects, they
still occ#r at different times (for, occurring at a different time is a
necessary condition of y's being a "repetition* of x). Thus, I can say that
χ occurred at time t, but cannot truly predicate the same of y, and, conse-
quently, it would seem that χ is not identical with y.
This sort of reasoning, it seems to me, is based on a specific Interpreta-
tion of Leibniz's law. Moreover, I am not altogether certain that it is the
"correct* Interpretation. It is even possible, I think, that the formulation
of Leibniz's law might need to be slightly modified, in order to effectively
cope with a particular sort of question.
For example, the sheet of paper that I am now writing on has a cer-
tain set of properties. If I were to tear it out of my notebook, and then
were to wad it into a ball, not only would it (this wad of paper) occupy
a different time than the sheet of paper on which I was formerly writing,
it would also possess a completely different set of properties. Now the
question arises: Is this wad of papier identical with the sheet of paper on
which I previously wrote?
But even more relevantly, if Magnus is correct in asserting that
occurring at different tirnes in a series is sufficient to defeat identity, then
is the Magnus of today identical with the Magnus of five years ago? For
that matter, is he even identical with the Magnus of five seconds ago? If
Magnus has, in fact, been the same person throughout his life, then how can
this be reconciled with his claim that occurring at different times "defeats"
identity? What is it that distinguishes occurring at different times within
a Ufetime from occurring at different times within the series of recurrenccs
in virtue of which the former may involve persistent identity, while the
latter may not?
In conclusion, perhaps two events, χ and y, may legitimately be said
to be identical, even though they occur at different times* This assumption,
at any rate, would seem to eliminate the need for asking such questions s
the above*