Dear author,

Please note that changes made in the online proofing system will
be added to the article before publication but are not reflected in
this PDF.

We also ask that this file not be used for submitting corrections.
 TREE 2414 1–11

1 Opinion

2
What Conservation Does
3Q1 [TD$FIRSNAME]Laurent[TD$FIRSNAME.] [TD$SURNAME]Godet[1,3,
TD$SURNAME.] * and [TD$FIRSNAME]Vincent[TD$FIRSNAME.] [TD$SURNAME]Devictor[2,3
TD$SURNAME.]

New agendas for conservation are regularly proposed based on the ground that Highlights
existing strategies are overly pessimistic, restricted to biodiversity hotspots, We test how conservation documents
biodiversity status, threats, and
and inappropriate to halt biodiversity loss. However, little empirical evidence
solutions.
supports such claims. Here we review the 12 971 papers published in the
leading conservation journals during the last 15 years to assess what conser- Many threats to biodiversity are
reported as well as some taxonomic
vation actually does. Although conservation research is affected by specific bias.
bias, conservation is playing a major role in providing empirical evidence of
human impacts on biodiversity. Encouraging biodiversity comebacks are also However, biodiversity comebacks are
documented as well as effective con-
published and a wide range of conservation tools, beyond the development of servation tools.
protected areas in wilderness areas, are promoted. We argue that finding new
New routes to conservation are neither
routes to conservation is neither necessary nor sufficient to halt biodiversity
necessary nor sufficient to halt biodi-
loss. versity loss.

4 The Agenda of Conservation Science Discredited

5Q3 Biodiversity loss is not decelerating [1], either in the terrestrial [2] or marine biotas [3] and affects
6 most taxa, particularly birds and mammals [4]. In addition to its velocity, one of the features of
7 the current situation lies in its anthropogenic origin [5]. This critical transition has been labeled as
8 ‘Anthropocene’ by scientists, policymakers, and the broader public [6]. However, these
9 conditions were already recognized more than three decades ago, when they led to the
10 emergence of the discipline of conservation biology, designated as ‘crisis discipline’ [7], whose
11 concepts and practices revolved around a fundamental conflict between protecting nature and
12 the acceleration of human impacts on biodiversity [8].

13 However, the relevance of conservation agenda has subsequently been challenged by three
14 main line of arguments. The first one argues that accumulating evidence of negative impacts of
15 human activities is counterproductive [9]. According to this view, conservation conveys overly
16 pessimistic messages and the discipline should review its messages of despair and hopeless-
17 ness [10,11]. The second uses the ongoing negative trend in biodiversity loss to claim that
18 traditional conservation tools, and mainly protected areas (see [12]), are insufficient to slow
19 down the biodiversity crisis [13]. Finally, conservation is regularly undermined as being
20 restricted to (tropical) biodiversity hotspots instead of paying attention to places where most
21 humans live and work [11]. 1
CNRS, Université de Nantes, UMR Q2
LETG, B.P. 81223, 44312 Nantes
Cedex 3, France
22 Such criticism remains, however, rhetorical and uninformed by empirical analysis. This has 2
CNRS, Université de Montpellier 2,
23 generated skeptical environmental postures as popularized by the publication of Lomborg’s UMR Institut des Sciences de
24 book in 2001 [14] and the passionate debate it triggered [15–17]. Skeptical environmentalism l’Evolution, 34095 Montpellier Cedex
05, France
25 already suggested that environmental problems, including biodiversity loss, lead to overly 3
Both authors contributed equally to
26 pessimistic claims and ineffective policies and are correlated with poverty. Therefore, according this work
27 to this view, environmental challenges should mostly reduce to ensuring economic and social
28 development. As already discussed by Doak et al. [18], this ancient rhetoric is surprisingly
*Correspondence:
29 recently used again by proponents of ‘new’ conservation movements as a justification to define laurent.godet@univ-nantes.fr
30 new directions to the field. In particular, this proposal suggests that conservation biology should (L. Godet).

Trends in Ecology & Evolution, Month Year, Vol. xx, No. yy https://doi.org/10.1016/j.tree.2018.07.004 1
© 2018 Elsevier Ltd. All rights reserved.
 TREE 2414 1–11

31 endorse a more integrative approach (and thus be called ‘conservation science’) to combine
32 human development and biodiversity issues. Yet, if uncritically explored, this litany may create
33 false dichotomies (between old and new and/or conservation science and conservation
34 biology). Indeed, conservation was meant, in its origin, to be a synthetic and multidisciplinary
35 science involving natural and social sciences, and dealing with natural resource fields such as
36 public policy, management, or forestry [7]. Overall, whether conservation needs to be reframed
37 toward even more compromise between biodiversity protection and human activities should be
38 further explored and anchored in empirical investigations rather than limited to wishful thinking.

39 In their paper ‘What is the future of conservation?’, Doak et al. [18] have already critically
40 assessed the major arguments used by those pleading for the need to find new directions to
41 conservation. They convincingly show that this rather dramatic proposal rests on unchecked
42 and dubious assumptions (e.g., past conservation has been a failure, or conservation should be
43 aligned with people’s interest, i.e., economic issues). However, a systematic review of what is
44 achieved in academic research is missing in this recurrent debate. What have we learnt from the
45 thousands of papers that have accumulated in conservation journals? Is conservation biology
46 condemned to report biased biodiversity status? Are threats reported by conservation biol-
47 ogists exaggerated and solutions undermined? Here, we review each of the 12 971 papers
48 published from January 2000 to February 2015 in the nine leading conservation science
49 journals, in order to investigate how conservation contributes to produce knowledge regarding
50 the status and threats of biodiversity and the solutions to its loss.

51 Identifying the Status, Threats, and Solutions Considered by Conservation

52 Science
53 Scrutinizing the Academic Conservation Corpus
54 We analyzed the titles, abstracts, and, if necessary, full-texts of all scientific papers published
55 from January 2000 to February 2015 in the nine leading international conservation science
56 journals. We restricted this analysis to journals corresponding to the field ‘conservation’
57 according to the Web of Science: Animal Conservation, Biodiversity and Conservation, Bio-
58 logical Conservation, Conservation Biology, Conservation Letters, Diversity and Distributions,
59 Environmental Conservation, Journal for Nature Conservation, and Oryx (n = 12 971 papers).
60 See Supplemental Material S1 online for details of the journal selection process.

61 Defining Status, Threats, and Solutions

62 We distinguished three categories of dominant justification to each paper: those dealing with
63 biodiversity ‘status’, those exploring ‘threats’ to biodiversity, and others proposing ‘solutions’
64 to biodiversity issues. We focused on paper using either empirical data or meta-analysis. Other
65 papers, not exploring one of the aforementioned categories or not dealing with biodiversity
66 issues (e.g., about conservation funding, new conservation policies), were classified as ‘other’.

67 The ‘Status’ category was defined as papers that report the state of biodiversity, without any
68 direct causal link with a specific threat.

69 The ‘Threats’ category was defined as papers that report threats to biodiversity (i.e., studies
70 which identify a causal link between a pressure and any biodiversity component).

71 The ‘Solutions’ category was defined as papers that report the influence of any conservation
72 tool on biodiversity. Papers proposing fuzzy conservation measures that cannot be used
73 directly by practitioners (such as ‘large patches of forests need to be protected’, or ‘fragmen-
74 tation has to be avoided’) were included in the ‘other’ category.

2 Trends in Ecology & Evolution, Month Year, Vol. xx, No. yy

 TREE 2414 1–11

75 The other category also included papers reporting equivocal status of any biodiversity com-
76 ponent (e.g., landscape fragmentation has positive effects on one taxon but negative effects on
77 another), methodological papers, descriptive papers (e.g., species ecology), opinions, episte-
78 mological studies, perspectives in conservation biology, development of indicators, theoretical
79 studies, improvement of knowledge of biodiversity, and editorials or book reviews.

80 Papers classified in the three main categories (status, threats, and solutions) were then
81 assigned to a finer and hierarchical classification. We distinguished four types of status (good
82 status, potential good status, bad status, and potential bad status, and then subdivided them
83 into different subcategories); 10 types of threats (subdivided into 20 different subcategories);
84 and 14 types of solutions (subdivided in four subcategories: failure, insufficiency, proposal, and
85 success). Note that the three main categories (status, threats, and solutions) were defined a
86 priori, but the subcategories emerged a posteriori from the initial reading of the papers, and
87 were gradually modified. The classification was checked twice, independently by the two
88 authors of this paper, to confirm the match between papers and categories. The exact
89 description of each category and subcategory is given in Supplemental Material S2 online.

90 In addition to this typology, various additional characteristics of each study were distinguished.
91 Geographical location: continents (Africa, Antarctica, Asia, Europe, North America, South
92 America, and Oceania); oceans/seas (Arctic Ocean, Atlantic Ocean, Baltic Sea/North Sea,
93 Caribbean Sea, Pacific Ocean, Indian Ocean/Persian Gulf, and Mediterranean Sea); and other
94 (diverse continents/oceans/seas, worldwide, and ex situ studies). Mexico and Central America
95 were included in South America. Ecological system considered: single species, multispecies,
96 local ecosystems, or global scale. And finally, the taxonomic group studied: amphibians, birds,
97 mammals, fish, freshwater invertebrates, marine invertebrates, terrestrial non-insect inverte-
98 brates, insects/arachnids, reptiles (excluding marine turtles), marine turtles, plants, lichens,
99 fungi/bryophyte/algae, mix of species, and other species.

100 Diversity and Distribution of Status, Threats, and Solutions in Space and
101 among Taxonomic Groups
102 From January 2000 to February 2015, the nine journals published 12 971 papers. The annual
103 publication rate increased from 2000 to 2014, partly due to the appearance of three new
104 journals: Conservation Letters, Diversity and Distributions, and the Journal for Nature Conser-
105 vation. The year 2015 was not included in this analysis as data were not available for the full year
106 (linear trend: +35 papers a year, F1,13 = 49.25, P < 0.01, R2 = 0.79).

107 The analysis showed that 76% of papers (9844) could not be classified in the three main
108 categories (status, threats, or solutions) and were therefore included in the ‘other’ category. Of
109 the 3127 remaining papers, 34% documented the specific status of a biodiversity component
110 (among which 125 and 767 papers were, respectively, showing ‘bad’ and ‘potentially bad’
111 status; 147 and 11 papers ‘good’ and ‘potentially good’ status); 31% reported threats to
112 biodiversity [among which most of the papers were related to habitat changes (352), overex-
113 ploitation of the resources (217), and the impact of invasive species (133)]; and 35% reported
114 solutions [mostly related to protected areas (207), followed by sustainable conservation
115 measures (182), species translocation (154), and habitat restoration or creation (118)] – see
116 Figure 1.

117 Among the solutions, we can notice that proposals of solutions (in blue on Figure 1) not only
118 dominated the studies but also that most of the papers proposed a sustainable use of the
119 resources (133), followed by species-specific management (97), species translocation (95), and

Trends in Ecology & Evolution, Month Year, Vol. xx, No. yy 3

 TREE 2414 1–11

Status (1050 papers)

Family of category Category

Bad state Bad state 125
125
Potenal bad state
Potenal bad state 767
767
Good state Spontaneous comeback 87
147 Biodiversity resistance-resilience 60

Potenal good state Potenal good state 11

11
0 50 100 150 650 700

Threats (983 papers)

Family of category Category
Habitat change Habitat degradaon 263

352 Habitat destrucon 89

69 Animal hunng
61 Fishing
Overexploitaon 53 Trading
217
19 Poaching
15 Earth resource exploitaon
Invasive species
Invasive species 133
133
27 Human disturbance

Human selements and presence 27 Human selements (roads and railways)

20 Human selements (excluding roads and railways)
100
19 Conflicts between humans and nature

7 Conflicts between humans

50 Agriculture
Agriculture, forestry, aquaculture
8 Forestry
60
2 Aquaculture
19 Genec limitaon
Ecological 15 Catastrophic natural events
47 8 Other ecological causes
55 Predaon
Diseases
32 Diseases
32
Climate change
20 Climate change
20
Polluon
18 Polluon
18
Human abandonment
44 Agricultural land abandonment
4
0 50 100 150 200 250

Soluons (1094 papers)

Family of category Category

Protected areas 18 40 59 90 207

Sustainable use 5 133 44 182

Translocaon 11 95 48 154

Habitat restoraon–creaon 3 1 85 29 118

Species-specific management 3 97 10 110

Idenficaon of hot areas for conservaon 70 70

Control of invasive species or pathogens 1 37 24 62

Landscape/seascape management 53 53

Species protecon 21 40 43 Failure (46)

Limitaon of human acvies 43 43

Insufficiency (42)
Social or economic 26 26
Proposal (718)
Other or mix of measures 11 11
Success (288)
Ex situ conservaon 343 10

Policy 5 5

0 50 100 150 200 250

Q5 Figure 1. Number of Papers within Each Category of Status of Biodiversity, Threats on Biodiversity, and
Solutions to Conserve Biodiversity.

4 Trends in Ecology & Evolution, Month Year, Vol. xx, No. yy

 TREE 2414 1–11

120 habitat restoration or creation (85), while the proposal of solutions related to protected areas
121 (namely extension or creation of protected areas) gathered only 59 papers. Moreover, focusing
122 only on those studies demonstrating the insufficiency of conservation solutions (in black on
123 Figure 1), we found that most papers conclude that protected areas are not enough (40 papers
124 among the 207 papers on protected areas), while the insufficiency of other types of solutions
125 was much less numerous.

126 Mapping the study sites of all 3127 papers revealed their uneven geographical distribution
127 (Figure 2). Only 10% focused on oceans and seas, compared with 85% of terrestrial studies.
128 The remaining 5% examined both terrestrial and marine areas, or were not linked to a particular
129 area. Focusing on papers assessing the status of biodiversity, we found that studies conducted
130 in Europe were the most numerous (64), followed by Asia (46) and Africa (46), North America
131 (35), South America (32), and, far behind, Oceania (9). More studies documented a good status
132 of biodiversity in Europe, Africa, and South America, contrary to the other continents. Focusing
133 on papers reporting threats to biodiversity, a comparable number of studies were located in
134 Asia (182), Europe (160), South America (154), North America (148), while only 64 were located
135 in Oceania. Wherever the studies were conducted, habitat change remained the major threat
136 reported, either followed by the overexploitation of resources, the impact of invasive species, or
137 the influence of human settlement–presence (in slightly different proportions). Finally, most of
138 the papers dealing with conservation solutions were related to study sites located in Europe
139 (296), followed by North America (178), Asia (132), Africa (127), Oceania (108), and South
140 America (104). Wherever the studies were conducted, proposal of solutions gathered the
141 largest number of papers, followed by papers documenting success, and, far behind, those
142 documenting failures and insufficiencies.

143 If most of the papers documenting a conservation status (75%) or conservation solutions (55%)
144 were focused on a single species, papers related to different species dominated the literature
145 related to threats to biodiversity. Looking at the single-species studies, most of the papers
146 focused on birds and mammals (Figure 3).

147 The Diversity and Effectiveness of Conservation Solutions

148 Our review reveals a wide range of conservation solutions, with numerous types of solutions
149 ranging from traditional conservation tools such as protected areas and rather ‘technical’
150 solutions (species translocation, habitat restoration–creation, species-specific management,
151 etc.), to ‘human-friendly’ measures such as promoting a sustainable use of natural resources,
152 or social and economic solutions rather than restricting conservation strategies to nature
153 protection. If protected areas remain the most documented solutions, studies promoting a
154 sustainable use are taking the second place and even the first place if we only focus on the
155 proposals of solutions. Contrary to what is advocated by new conservationists, it thus suggests
156 that human-friendly solutions are already commonly proposed by the scientific literature.

157 There is probably a bias toward reporting or not reporting failures of conservation tools,
158 although we are unable to document in what direction our data can be biased. We may
159 expect that a failure of a conservation tool is likely to be more reported than a success because
160 the latter is expected and may be less appealing to journals. At the opposite, we also may
161 expect that conservation biologists are reluctant to document conservation failures because it
162 is potentially much more controversial. Moreover, it is probably more difficult to publish a study
163 reporting no effect of a conservation tool (what corresponds to a failure), as this may corre-
164 spond either to a real inefficiency or to an impossibility of showing significant effects with the
165 data used. Considering this, we identified few studies reporting failures (only 46), against 288

Trends in Ecology & Evolution, Month Year, Vol. xx, No. yy 5

 TREE 2414 1–11

Status
N

Europe (64)

Asia (46) Bad status

North
America
(35)
Good status
Region (number of papers)
Carib.
Sea (8) Africa (46)
Atlan c
Pacific Ocean (8) Ocean (4)

Indian
Ocean (8)
Oceania
(9)
South
0 4000 km America
(32)

Threats
N

Europe (160)

North Asia (182) Habitat change

America
(148) Overexploita on
Invasive species
Carib. Med. Sea (10)
Atlan c Africa Human se lement–presence
Sea (132)
(12)
Ocean (19) Agriculture aquaculture and
Pacific Ocean (41) forestry
Indian Ecological
Ocean (15) Oceania
(64) Disease
Other
0 4000 km South
America Region (number of papers)
(154)

Solu ons
N

Europe (296)
Failure
North
America
Insufficiency
(178) Asia (132)
Proposal
Success
Carib. Med. Sea (10) Region (number of papers)
Sea
Atlan c Africa
(9)
Pacific Ocean (37) Ocean (127)
(19)
Indian
Ocean (20) Oceania
(108)
South
0 4000 km America
(104)

Figure 2. Spatial Distribution of Papers Reporting Status of Biodiversity, Threats on Biodiversity, and Solutions to Conserve Biodiversity. Carib. Sea,
Caribbean Sea; Med. Sea, Mediterranean Sea.

6 Trends in Ecology & Evolution, Month Year, Vol. xx, No. yy

 TREE 2414 1–11

Status
Mammals

Birds

plants

Amphibians
75%
Reples (excluding marine turtles)

Insects and arachnids

Marine turtles

Fish

Non-insect terrestrial invertebrates

Marine invertebrates

24% Freshwater invertebrates

Fungi, bryophytes, algae

0 50 100 150 200

Threats
Mammals
Birds

35% Reples (excluding marine turtles)

Plants
Amphibians
Insects and arachnids
Fish
Marine invertebrates
Freshwater invertebrates
52%
Non-insect terrestrial invertebrates
Mulspecies
Lichens
Fungi, bryophytes, algae
10%
0 50 100 150 200

Soluons
Mammals

Birds

Reples (excluding marine turtles)

55% Plants

Amphibians

Insects and arachnids

fish

Marine invertebrates

30% Freshwater invertebrates

Non-insect terrestrial invertebrates

Lichens
8%
Fungi, bryophytes, algae
7%
0 50 100 150 200
Species
Mulspecies
Habitats
Biodiversity and earth

Figure 3. Proportion of Papers Included in Status of Biodiversity, Threats on Biodiversity, and Solutions to
Conserve Biodiversity, According to the Group Considered, and, for Papers Dealing with Only One species,
Number of Papers by Taxonomic Group.

Trends in Ecology & Evolution, Month Year, Vol. xx, No. yy 7

 TREE 2414 1–11

166 reporting successful measures. A total of 90 studies reported the success of protected areas,
167 18 their failures, and 40 their insufficiency, suggesting that although protected areas are not a
168 panacea, they still represent a key strategy for the success of specific conservation goals.
169 Moreover, conservationists seem to be aware and critical toward the issues protected areas
170 may raise rather than promoting this tool blindly.

171 Diversity in Study Areas and Species

172 We found great diversity in focus regarding geographical areas and species. As it has been reported
173 elsewhere (e.g., [19–23]), more studies have been conducted in terrestrial than marine systems.
174 Most studies were species specific, and essentially focused on birds and mammals. This taxa bias
175 was demonstrated by Amori and Gippoliti [24]. It continues to be used to justify the revision of
176 conservation agendas, specifically to take the focus off charismatic species and biodiversity
177 hotspots, and instead pay attention to the rest of the planet, including places where humans live
178 [11]. Our review partly challenges this view. Most studies are conducted in European developed
179 countries, far from tropical biodiversity hotspots. Moreover, these biases provide support for even
180 more effort being given to broadening, rather than restricting the scope of conservation.

181 Realistic Conservation: Beyond Pessimism and Optimism

182 Investigating the past, present, and future of conservation science needs empirical support.
183 Assessing its effectiveness should rest on empirical studies that make an objective evaluation of
184 the failure and success of conservation actions [25–27]. Long-term progress toward the global
185 objective of reducing biodiversity loss has led to massive efforts dedicated to the inventory,
186 survey, and monitoring of biological components in space and time. Lists of biodiversity
187 indicators (e.g., the Red List [28] or the Living Planet Index [29]) are regularly updated by
188 work that is supported by the Convention on Biological Diversity [30]. Nevertheless, we lack a
189 clear overview of what knowledge is actually produced by conservation science. Therefore, our
190 objective was to paint a general picture of what conservation actually does in terms of
191 knowledge production related to the status, threats, and solutions for the biodiversity crisis.

192 Overall, our review is consistent with a realistic picture, which is neither overly pessimistic nor
193 optimistic. By ‘realistic’ we of course do not mean that conservation science is free from any bias or
194 depicts the ‘reality’ regarding the state of biodiversity. On the contrary, we document clear uneven
195 distributions of research efforts in space and among taxonomic groups. Besides, conservation
196 biology is not meant to substitute other and complementary initiatives tracking biodiversity trends:
197 the global and regional trends of biodiversity loss are not primarily reported by conservation studies
198 as those considered in our paper but rather by constant and regularly updated efforts of
199 environmental nongovernmental organizations [e.g., World Wildlife Fund (WWF) or International
200 Union for Conservation of Nature (IUCN) reports]. We found that in academic journals, many local
201 and independent but empirically supported papers are still documenting threats, and among the
202 papers documenting a status of a biodiversity component, many report a potential bad state. This
203 highlights the accumulation of empirical evidence of degradation. Similar findings were revealed by
204 Bini et al. [31] in 2004, suggesting that there is no sign of significant improvement in the more than
205 10 years that have passed. We also found that the main threats were, in decreasing order
206 generated by, habitat change, overexploitation, and invasive species. This picture is consistent
207 with the well-documented ‘evil quartet’ [32] of the main threats to biodiversity, and cannot be
208 considered as resulting from the failure of conservation concepts and actions per se.

209 Whether, and how, conservation science should communicate bad news is controversial.
210 Garnett and Lindenmayer [9] argue that developing a narrative focused on mass extinction
211 through the accumulation of bad news could be counterproductive. Recently, specific

8 Trends in Ecology & Evolution, Month Year, Vol. xx, No. yy

 TREE 2414 1–11

212 proposals, such as ‘green lists’ and reporting good news, have been proposed. Authors who
213 plead for the development of ‘a new conservation science’ go further, and consider traditional
214 conservation as overly pessimistic [11,33]. They advocate the need to endorse a ‘more
215 optimistic, human-friendly vision’ [11], and to recognize that nature is resilient [33], even when
216 faced with the most powerful human disturbances [11]. By contrast, others have argued that
217 focusing on good news could obscure the need to adapt to major changes, and fail to
218 acknowledge the conflict between human activities and the need for nature protection [34].
219 Our review suggests that aside from any personal or ideological position, there is no particular
220 reason to downplay the worrying messages delivered by conservation sciences [35], or ignore
221 the situation is degrading worldwide (e.g., [36]).

222 The pessimistic view is however partly counterbalanced by the reports of good status of
223 biodiversity components we found. Actually, focusing on the status regardless of the threats,
224 we even found a larger number of papers documenting a good status than a bad status in different
225 areas (Europe, Africa, and South America). The progressive return of wildlife is, for example,
226 documented in Europe [37] where the comeback of large carnivores (wolf, brown bear, lynx, and
227 wolverine) highlights that the protection and welcoming of large species is possible [38]. Our study
228 also highlighted the very large number of proposals of solutions. These studies argue that the
229 application of the methods they develop could enhance the future state of biodiversity.

230 Finally, the large number of papers classified as ‘other’ (76%) also reveal that conservation
231 biology, as a discipline, tends to develop knowledge and methods related to conservation
232 issues, rather than making clear statements with respect to biodiversity status, threats, or
233 conservation solutions. Among those papers, many opinion papers or epistemological studies
234 also contribute to generate important debates, contradictory arguments, and reflexive view-
235 points. These, far from reflecting a useless scientific research, are rather signs of good scientific
236 activity. Although it is beyond the scope of this paper, note that the ‘other’ category gathers
237 different subcategories that could be explored deeper in a future bibliographic analysis. It could
238 be relevant to distinguish on the one hand the papers which can be considered as irrelevant for
239 our analysis (e.g., not dealing with biodiversity issues, methodological papers, descriptive
240 papers, opinions, epistemological studies, perspectives in conservation biology, development
241 of indicators, theoretical studies, improvement of knowledge of biodiversity, and editorials or
242 book reviews) and on the other hand the papers related to insufficient information (e.g.,
243 equivocal status or fuzzy conservation measures).

244 Note that our framework is reflecting the inherent biases affecting scientific publication. For
245 instance, stable biodiversity status, harmless anthropic impacts, and ineffective conservation
246 solutions are probably more difficult to publish. The extent to which including those overlooked
247 studies would change our conclusions is difficult to know. However, conservation legitimacy
248 and credibility would clearly gain from higher transparency in the publication of what is working
249 or not. We think that evidence-based conservation is a good step in that direction [39]. Access
250 to unpublished reports or promoting higher visibility of unexpected conservation results should
251 also be facilitated. The need for evidence-based conservation was defended on the ground that
252 conservation practice is ‘based upon anecdote and myth rather than upon the systematic
253 appraisal of the evidence’ [39]. A specific database collating effective or ineffective actions can
254 increase the likelihood to find the best solution (or avoid the worth) to a given specific issue. This
255 would allow finer analysis on what contributes to conservation success and failure and their
256 respective timing. For instance, it is likely that for some specific groups a delay is necessary
257 between a given conservation action and the first positive signs. The extent to which this reveals
258 what problems may be more reversible in specific situations could then be scrutinized.

Trends in Ecology & Evolution, Month Year, Vol. xx, No. yy 9

 TREE 2414 1–11

259 Concluding Remarks Outstanding Questions

260 The aim of our study was to take a step back from current, simplistic, ideological positions, and The difficulty to address conservation
261 investigate the scientific conservation literature. In doing so, we do not claim to contribute to challenges does not result from the
conservation agenda itself. The new
262 classical ethical debates on conservation motives (e.g., as in the long-lasting discussions about conservation debate should focus on
263 wilderness and intrinsic values) or to more recent initiatives on global land sparing or sharing (e. what is working or not and why, rather
264 g., Nature Needs Half movement [40]). But our empirical approach provides sufficient argu- than proposing new directions for the
discipline. Recognizing the existence
265 ments to conclude that conservation biology does not accumulate studies dedicated to strict
of several biases in the distribution of
266 nature protection with no or little connexion to human well-being. Rather, we show that several research studies among countries,
267 targets including the protection of nature for itself, nature for people, and nature with people are taxonomic groups, or topics should
268 coexisting [8]. Interestingly, among most effective conservation actions, protected areas are still motivate even higher conservation
efforts. In the future, the success sto-
269 playing a major positive role. More generally, as the world becomes ever more dominated by
ries of conservation actions and the
270 humans, conservation of species in the wild is in fact even more relevant [41]. What we argue, effectiveness of existing conservation
271 however, is that proposing dramatic, top–down changes to a discipline, or accusing it of tools should be better investigated.
272 ‘failure’, not only appears counterproductive, but also ignores the dynamics of science itself.
273 But more importantly, this debate is a classical diversion from the more profound causes of
274 biodiversity loss. We thus concur with adopting more lucid approaches to existing and non-
275 reducible conflicts between biodiversity conservation and the deny of any limits to destructive
276 human activities. Calling for new directions to conservation is neither necessary nor sufficient.
277 Constantly seeking even more compromise when nature protection is at stake is part of the
278Q4 major obstacles, not of the solutions (see Outstanding Questions).

279 Supplemental Information

280 Supplemental information associated with this article can be found, in the online version, at https://doi.org/10.1016/j.tree.
281 2018.07.004.

282 References
314
283 1. Leakey, R. and Lewin, R. (1996) The Sixth Extinction. Patterns of 18. Doak, D.F. et al. (2014) What is the future of conservation? Trends
315
284 Life and the Future of Humankind, Anchor Ecol. Evol. 29, 77–81
316
285 2. Dirzo, R. et al. (2014) Defaunation in the Anthropocene. Science 19. Kaufman, L. (1988) Marine biodiversity: the sleeping dragon.
317
286 345, 401–406 Conserv. Biol. 2, 307–308
318
287 3. Payne, J.L. et al. (2016) Ecological selectivity of the emerging 20. Irish, K.E. and Norse, E.A. (1996) Scant emphasis on marine
319
288 mass extinction in the oceans. Science 353, 1284–1286 biodiversity. Conserv. Biol. 10, 680
320
289 4. International Union for Conservation of Nature (IUCN) (2012) The 21. Kochin, B.F. and Levin, P.S. (2003) Lack of concern deepens the
321
290 IUCN Red List of Species, IUCN ocean’s problems. Nature 424, 723
322
291 5. Kolbert, E. (2015) The Sixth Extinction: An Unnatural History, 22. Levin, P.S. and Kochin, B.F. (2004) Publication of marine conser-
323
292 Picador, (USA) vation papers: is conservation biology too dry? Conserv. Biol. 18,
324
293 6. Crutzen, P.J. (2002) Geology of mankind. Nature 415, 23 1160–1162
325
294 7. Soulé, M. (1985) What is conservation biology? Bioscience 35, 23. Norse, E.A. and Crowder, L.B. (2005) Marine Conservation Biol-
326
295 727–734 ogy, Island Press
327
296 8. Mace, G. (2014) Whose conservation? Science 345, 1558–1560 24. Amori, G. and Gippoliti, S. (2000) What do mammalogists want to
328 save? Ten years of mammalian conservation biology. Biodivers.
297 9. Garnett, T.S. and Lindenmayer, D.B. (2011) Conservation science
329 Conserv. 9, 785–793
298 must engender hope to succeed. Trend Ecol. Evol. 26, 59–60
330 25. Chape, S. et al. (2005) Measuring the extent and effectiveness of
299 10. Kareiva, P. and Marvier, M. (2012) What is conservation science?
331 protected areas as an indicator for meeting global biodiversity
300 Bioscience 62, 962–969
332 targets. Philos. Trans. R. Soc. Lond. B Biol. Sci. 360, 443–445
301 11. Kareiva, P. et al. (2012) Conservation in the Anthropocene:
333 26. Butchart, S.H.M. et al. (2006) How many bird extinctions have we
302 beyond solitude and fragility. Breakthrough J. 2, 29–37
334 prevented? Oryx 40, 266–278
303 12. Kareiva, P. (2014) New conservation: setting the record straight
335 27. Rodrigues, A.S.L. (2006) Are global conservation efforts success-
304 and finding common ground. Conserv. Biol. 28, 634–636
336 ful? Science 313, 1051–1052
305 13. Marvier, M. and Kareiva, P. (2014) The evidence and values
337 28. Collar, N.J. (1996) The reasons for red data books. Oryx 30, 121–
306 underlying ‘new conservation’. Trends Ecol. Evol. 29, 131–132
338 130
307 14. Lomborg, B. (2001) The Skeptical Environmentalist: Measuring
339 29. Loh, J. et al. (2005) The Living Planet Index: using species
308 the Real State of the World, Cambridge University Press
340 population time series to track trends in biodiversity. Philos.
341 15. Pimm, S. and Harvey, J. (2001) The skeptical environmentalist: Trans. R. Soc. Lond. B Biol. Sci. 360, 289–295
309 measuring the real state of the world. Nature 414, 149–150
342 30. Balmford, A. et al. (2005) Science and the Convention on Biolog-
310
343 16. Moowaw, W.R. (2002) Lomborg’s The skeptical environmentalist: ical Diversity’s 2010 target. Science 307, 212–213
311 refuting a scientific model without science. Conserv. Biol. 16,
344 31. Bini, L.M. et al. (2004) Lomborg and the litany of biodiversity crisis:
312 861–862
345 what the peer-reviewed literature says. Conserv. Biol. 19, 1301–
313
346 17. Giles, J. (2003) The man they love to hate. Nature 423, 216–218 1305

10 Trends in Ecology & Evolution, Month Year, Vol. xx, No. yy

 TREE 2414 1–11

358
347 32. Diamond, J.M. (1984) ‘Normal’ extinction of isolated populations. 37. Deinet, S. et al. (2013) Wildlife Comeback in Europe: The Recovery of
359 In Extinctions (Nitecki, M.H., ed.), pp. 191–246, Chicago Univer- SelectedMammalandBirdSpecies – FinalReport to RewildingEurope
360
348 sity Press by ZSL, BirdLife International and the European Bird Census Council
349 33. Kareiva, P. and Marvier, M. (2007) Conservation for the people. 38. Chapron, G. et al. (2014) Recovery of large carnivores in
361
350 Sci. Am. 297, 50–57 Europe’s modern human-dominated landscapes. Science
362
351 34. Arlettaz, R. et al. (2011) Active scepticism must drive biodiversity 346, 1517–1519
363
352 conservation science. Trend Ecol. Evol. 26, 379–380 39. Sutherland, W.J. et al. (2004) The need for evidence-based
364
353 35. Johns, D. et al. (2017) Caring, killing, euphemism and George conservation. Trend Ecol. Evol. 19, 305–308
365
354 Orwell: how language choice undercuts our mission. Biol. Con- 40. Büscher, B. et al. (2017) Half-Earth or whole Earth? Radical ideas
366
355 serv. 211, 174–176 for conservation, and their implications. Oryx 51, 407–410
367
356 36. Estrada, A. et al. (2017) Impending extinction crisis of the world’s 41. Stokes, D.L. (2018) Why conserving species in the wild still
368
357 primates: why primates matter. Sci. Adv. 3, e1600946 matters. Biodivers. Conserv. 27, 1539–1544

Trends in Ecology & Evolution, Month Year, Vol. xx, No. yy 11