Navigazione – Mappa del sito

HomeNumeri59Ideal and actual inventories of b...

Ideal and actual inventories of biodiversity

Anouk Barberousse e Sophie Bary
p. 14-31


Generalmente si ritiene che la rilevazione e identificazione delle specie presenti in una determinata area fornisca un corpus di conoscenza di base che permette ai biologi di sviluppare pezzi di conoscenza ulteriori. Tuttavia, si rivela sorprendentemente difficile ottenere inventari biologici che soddisfino i criteri inerenti a tale conoscenza di base. Il nostro obiettivo in questo articolo è di mettere in luce come la pratica corrente di inventariazione biologica sia condizionata da varie limitazioni e potenziali pregiudizi; cosa che ci conduce a riconsiderare le funzioni degli inventari all’inizio del ventunesimo secolo. A tale scopo, presentiamo l’esempio degli inventari della fauna degli ambienti marini profondi dell’Oceano Pacifico e ci concentriamo su una fonte di pregiudizio: la perdurante influenza della cosiddetta “ipotesi azoica”, secondo la quale non ci sarebbe vita al di sotto dei 600 metri. Mostriamo come l’ipotesi azoica, nonostante sia stata rapidamente confutata, abbia fortemente influenzato la concezione della fauna degli ambienti profondi. Nelle conclusioni, guardiamo come i limiti economici influenzino le pratiche correnti di inventario biologico degli ambienti marini profondi.

Torna su

Testo integrale

1The detection and identification of the species living on a given area is usually supposed to provide a corpus of basic knowledge enabling biologists to develop further pieces of knowledge. However, it reveals suprisingly difficult to achieve biological inventories satisfying the criteria pertaining to such basic knowledge. Our aim in this paper is to highlight how the current practice of biological inventory is shaped by various constraints and potential biases. This leads us to re-consider the functions of inventories at the beginning of the twenty-first century.

2There is a sharp contrast between inventories as they are actually practiced and inventories as they are conceived of or imagined. In order to discuss this contrast, we begin by presenting the main features that are commonly required of ideal inventories, that is, inventories fulfilling all the functions biologists which they could fulfill (section 1). Ideal inventories have themselves a long history the main features of which we recall in order to better appreciate what current biologists expect from their inventories. Then we present the example of deep-sea fauna inventories in the Pacific Ocean (section 2). By recalling parts of the history of deep-sea inventories from the beginning of the nineteenth century, we show that actual inventories have been and still are far from obeying the rules of ideal inventories. We focus on one source of bias in this case: the lasting influence of the so-called “azoic hypothesis”, according to which there is no life under 600 meters. We show how the azoic hypothesis has strongly influenced the conception of deep-sea fauna even though it has been quickly refuted (section 2.1). At the end of the paper, we look into the implications of economical constraints on the current practice of deep-sea biological inventory and show a strong link between the surprising persistence of the azoic hypothesis and economical pressures to exploit living and mineral resources (section 2.2). Both combine to affect scientific research in the deep-sea domain.

1. Requirements on inventories of biodiversity and their transformation over time

3Our aim in this section is to review what is expected from an inventory of biodiversity. The basic goal is not only to learn which species live on a given area, but also to evaluate how abundant they are, that is, whether many specimens are present or not. At first, this seems a mainly descriptive task, but as it involves species identification, it is also informed by the theoretical background underlying biological research at a given time – today, this background consists in the theory of evolution. This calls for an investigation of the contribution of biological inventories to biological knowledge at large.

4Another reason why we want to make explicit how inventories participate in the constitution of biological knowledge is that biologists of different disciplines adopt different images of how the various pieces of biological knowledge combine together. Indeed, the details of the image of biological knowledge slightly vary according to whether you are a taxonomist or not. The non-taxonimists usually consider the inventory task as a relatively simple one, or at least a simpler task than their own ones, which may involve laboratory work, long experiments, data difficult to interpret, etc. This view is extremely powerful. It associates species description with a sort of epistemic passivity, as the taxonomists are supposed to “just”, or “simply” state what they observe. However, this view of the role of taxonomists, and, correspondingly, of biological inventories, is misleading.

5This is why we begin by presenting how inventories of biodiversity participate in the development of what we call the “Web of Biological Knowledge” (WBK). It consists in the hierarchical body of interrelated pieces of biological knowledge and hypotheses at one time. By this name, we want to emphasise that empirical biological knowledge is always interrelated with theoretical hypotheses, even though there is a perceived hierarchy according to which knowledge coming from observation and experiment is conceived as the underlying basis for hypotheses and models. This body of knowledge evolves as time goes and is affected by the transformation of theories and practices. The major alteration of the WBK was brought about by the Darwinian revolution, which entirely transformed the set of admissible explanations in biology. The functions inventories of biodiversity fulfill within the WBK define a set of criteria inventories have to satisfy. In section 1.1, we present our view of the WBK. Then we go to the criteria inventories have to satisfy. In section 1.3, we put forward some general reasons why ideal inventories cannot be actually realised.

1.1. The Web of Biological Knowledge

6Because the analysis of the practice of biological inventory is doomed to rely on a vast aggregate of elements, both epistemic and material, we need to introduce a notion referring to the intellectual basis of this scientific practice, however difficult to capture it may be. The WBK is indeed the loosely structured set of beliefs, hypotheses, well-justified pieces of knowledge, theories, and epistemic tools, methods and techniques that constitutes what biologists have in common at a given time. It is usually difficult to precisely identify the WBK at a given time, and its content may vary according to the biologist one picks up. However, it seems to be a fair idealisation to assume the existence of such a set of beliefs and other epistemic elements. It is commonly considered that firmly-justified elements and theories are important to delimit a scientific field. By contrast, we want to emphasise that less-justified beliefs and assumptions also play a role in scientists’ mind and are therefore important to characterise the state of a discipline at a given time.

7In order to assess the relationships between biological inventories and the WBK, it is important to emphasis that it is usually difficult to clearly distinguish between well-justified elements within the WBK and less-justified ones, although the WBK is precisely characterised by both elements being closely mixed. Moreover, the WBK although includes methods and techniques that are partially independent from its theoretical framework, as we shall emphasise below.

8The hierarchical organisation of the WBK is twofold. First, there is an axis going from empirically-justified elements to more theoretical ones. Second, there is an axis of explicitness, going from the consciously accepted elements to hidden assumptions. The hidden assumptions may be unjustified but they can nevertheless influence how research is carried out.

1.2. Ideal inventories

9Discovering and describing the various species living on Earth, as well as the extinct ones, has been a major goal of naturalists since the eighteenth century at least, not to mention medieval herbals. From Linnaeus’ work onward, this task has been conceived as participating in a larger epistemic enterprise aiming at learning more about the living world. For sure, this enterprise was not only an epistemic one, as it was also attributed religious goals. Linnaeus himself explicitly declared, at the beginning of the Systema naturae, that his work was intended to mirror the structure of God’s creation. Our aim in this section is to present how inventories of fauna and flora are articulated to the more general epistemic context of biological knowledge available at a time and its transformations over time. We begin with Linnaeus’ time and show how the demands on biological inventories evolved with the transformation of biological knowledge.

10Inventories of fauna and flora are of two sorts. The first are oriented toward immediate agricultural or economical purpose whereas the second are governed by epistemic goals. For sure, the distinction cannot be strict; however, it helps determining the characteristic features of inventories that are at least partly detached from practical purpose. They have to satisfy a first, powerful requirement: systematicity. As many animal and plant species are not immediately visible, contrary to those whose size or agricultural importance makes salient, systematic exploration is required. For sure, systematic exploration of fauna and flora may have interesting agricultural outcomes, but from the modern times onward, it has been carried out without an eye on these outcomes, that is, for itself. More precisely, the ultimate goal of this slow and minute task was to better know what the Earth is populated of. From our contemporary perspective, we can detect that this task was not conceived as strictly epistemic in our sense, for discovering what the Earth is populated of was tantamount to recognising and honouring what God had put onto it. However, this religious side of the inventory task did not go against the systematicity requirement, but rather strengthened it.

11Together with systematicity, another requirement was efficiently put forward by Linnaeus: inventories of fauna and flora are to participate in a common epistemic enterprise. The rules of Linnaean nomenclature indeed guarantee that wherever you are, you have to use the same species names, and moreover to adopt the same rules for creating new species names if you happen to discover a specimen that is related to no already-known names. In this way, the Linnaean framework provided naturalists with the first unifying framework for biological inventories. This framework is not a theoretical one but is based on a set of practical rules that allow for the constitution of descriptive knowledge. We identify it as the first, common framework efficiently organising the WBK. Indeed, even though it is possible to characterise former attempts at biological knowledge as being potentially cumulative and universal, the Linnaean framework is the first one to put forward explicit, international rules for the constitution of a body of biological knowledge that can be accessed and enriched by anyone interested in the description of her local fauna and flora.

12The Linnaean framework for species description has to be clearly distinguished from the theoretical structure organising the WBK at a given time. As is well-known, fixist, creationist, and essentialist assumptions were commonly accepted at Linnaeus’ time. By contrast, our current WBK is organised by an evolutionary-theoretical structure. However, this difference does not imply that the Linnaean tool for nomenclature has to be rejected. Even though it has developed within a theoretical context that has been entirely superseded by evolutionary theory, it is still considered a valuable tool, enabling current taxonomists to carry on their work. At Linnaeus’ time, it already participated in an efficient dynamics of scientific knowledge production, and it still does. By following Linnaeus’ guidelines, it was possible to achieve genuine scientific knowledge, by the criteria of the time. Today, our criteria have changed, but the Linnaean tool is still adapted. Even though they have often been criticised, the Linnaean rules as still used as international standards. As a result, from Linnaeus onward, taxonomy has become a truly scientific field of knowledge.

13Let us briefly recall the principles of Linnaeus’ taxonomic procedures as they are carried out today. A species name is associated with a specimen in the following way. The starting point is the observation of some organisms, either living or fossils. Taxonomists then use different criteria in order to find out how many species they are members of. These criteria may be morphogenetic, physiological, ecological, or genetic. Whereas some of the species they thus identify are already known and described, they may also hypothesize that other specimens belong to unknown species. Then they have to give names to these previously unknown species, according to the rules gathered in the current nomenclature codes. The main feature of these rules is that a species name must be associated with a material specimen, to be found in a collection of natural history (this specimen is called a “holotype”). This association guarantees that biologists can always come back to the holotype in order to verify whether the description of the species that has been published based on the specimen is satisfying or has to be revised.

14In the same way as the Linnaean nomenclature tools can be articulated with different WBKs, the expectations associated with biological inventories may also change in relation to the transformation of the WBK over time. At Linnaeus’ time, naturalists worked toward the combined realisation of two goals: the establishment of an exhaustive catalogue of the living world for the glory of God and the elaboration of a rational classification. Today, we conceive of these two goal as entirely different, because, as emphasised above, biological inventories are integrated within a different theoretical structure. However, rational classification and, albeit to a lesser extent, systematic exploration are still conceived as main elements of the constitution of biological knowledge.

15There is an other major difference between Linnaeus’ time and today, which is closely linked with evolutionary theory being our underlying biological theory. Whereas taking a biological inventory could be conceived as a finite task, at least in principle, in the eighteenth century, we know that exhaustivity cannot possibly be the aim of current inventories. There are two reasons to this change. The first one is fundamental: species are conceived as evolving entities, so that there is no sense in trying to describe them once and for all. Inventories have thus acquired a temporary value; they cannot possibly be meant to hold for all times. The second reason is historical: we are now experiencing a new mass extinction which transforms any current inventory in a race against time. Moreover, we know that we are doomed to loose this race by far. Completeness has thus become but a relative notion. As a result, the very meaning of biological inventories has been radically transformed. Today’s inventories are more about characterising the structure of biodiversity in some geographical region as well as the processes that lead to this diversity than about writing an exhaustive catalogue expected to hold for all times.

16Inventories of biodiversity in the new sense we have just presented are still considered a precondition for the elaboration of further biological knowledge, even though they are sometimes judged less important than laboratory work. As a result, a doubly hierarchical image of biological knowledge has emerged, framing the WBK as we know it today. Within this hierarchical image, the information gained from systematic inventories is both basic, as it is the foundation of more sophisticated knowledge, and simpler in nature than the other parts of this epistemic building. According to the first axis we have identified in section 1.1, inventories allow for empirically well-justified knowledge because it is usually viewed as purely descriptive knowledge, in the sense that it is supposed to involve no hypothesis or theory. According to the second axis, it is supposed to be on the totally explicit side, because nothing hidden seems to be included in taxonomic descriptions.

17For sure, taxonomists do not share all assumptions underlying the hierarchical image we have just presented. They insist, by contrast, on the fact that their work is no less hypothesis-based than laboratory work. They indeed put forward that, rather than passively describing what they see, they offer hypotheses about the structure of biodiversity and its history. For instance, they make clear that a species name is first and foremost a label allowing them to refer to a hypothesis, according to which the associated organisms actually form a species, that is, will not converge again with other groups in the future. As other scientific hypotheses, taxonomic hypotheses are revisable, and are actually often revised.

18Even when one takes into account the dynamic of taxonomic knowledge and the hypothetical nature of species identification, it remains that taxonomic procedures themselves involve an implicit hierarchy with respect to epistemic actions. For instance, collecting specimens is assumed to be a simpler task than describing them in a publication. Moreover, this task is usually taken to be neutral vis à vis any scientific hypothesis, or, to put it in other terms, unaffected by any prevailing scientific theory. By contrast, the description task involves knowledge of the specimen’s phylogeny.

19Let us indicate what the perceived hierarchies that frame our current WBK imply. At the basis of further biological knowledge, taxonomic knowledge is meant to mostly consist in factual elements and to contain as few hypotheses as possible. By contrast, the further knowledge it is supposed to enable consists in the testing of phylogenetic or ecological hypothesis. More generally, it is based on inferences made from the facts revealed by the inventory. This is why the knowledge established by the inventory is usually conceived as purely descriptive, or hypothesis-free. According to this conception, which is close to the Vienna circle conception of scientific observation, a description is an intellectual action consisting in putting into words what perception provides the brain with. The basic requirement for this action is that the transformation of the object of perception into words should involve as few elements coming from the agent’s mind as possible. The agent should refrain from making any interpretation but rather set the facts as neutrally as possible.

  • 1 Cf. Carvalho et al. 2013.

20To sum up, within the global image of biological knowledge that has emerged since Linnaeus’ time, inventories of biodiversity occupy a special place. They are supposed to provide biologists with systematic and hopefully exhaustive knowledge of the organisms in a given area and this knowledge is supposed to be basic, that is, descriptive, neutral or factual. Even though taxonomists are clear that their work involves the elaboration of hypotheses1, and that attribution of a name to a species is provisional on the availability of further knowledge, the assumption remains that the knowledge they provide is more basic than other types of biological knowledge. At least, the part of biological knowledge deriving from the practice of inventories is supposed to be more factual than other parts.

21This global image of biological knowledge relies on a highly idealised notion of a biological inventory. As a matter of fact, no actual inventory is truly systematic and exhaustive and all of them include hidden assumptions that deviate them from basic descriptions of the organisms living on a given area. In the next session, we review some general reasons why actual inventories are different from their idealised notion. This will allow us to emphasise in section 2 that many other specific reasons can also hinder the realisation of inventories.

1.3. Some reasons why ideal inventories cannot be realised in practice

22After having presented the main features of ideal inventories, namely systematicity, exhaustivity, and theoretical neutrality, we now review some reasons why most inventories do not meet these features. Our prospect is to identify general reasons, that is, reasons that bear on most inventories and explain why they do not fulfill the above requirements. The general reasons we identify in this section are such that usually, there is hardly anything to do against them. Taxonomists are nevertheless aware that they hinder their enterprise. In section 2, we contrast these general reasons with other reasons that are both specific to our case-study and less easy to identify and to become aware of.

23Even though some exhaustive inventories can be, and have been, carried out, as the inventory of living birds, most inventories are not exhaustive or systematic enough to satisfy the above-presented ideal criteria. This may be for lack of time or lack of accessibility, as illustrated by marine inventories, and even more so by deep-sea inventories. Lack of time and of other resources, like money and instruments, e.g., fine dredgers, submersible crafts, high-pressure camera…, make up a major practical obstacle to the realisation of ideal deep-sea inventories.

24As far as the program of deep-sea fauna inventory in the South-West Pacific Ocean we will focus on in section 2 is concerned, the practical obstacles were partly removed in 1985, when the French research ship was at last equipped with a multi-beam swath bathymetry, allowing for both efficient collecting and depth measurement. As a result, researchers became suddenly aware of the heterogeneity of the deep-sea habitat as the topography of the deep-sea floor could be investigated in much more details than was previously possible. For instance, it became possible to differentiate subsea mounts from ridges and springs and to know on what kind of structure (seamount, ridge…) and at what depth the specimens were sampled. By contrast, without this equipment, accessibility was a major practical obstacle for deep-sea inventories. We put the above-mentioned obstacles into the category of general obstacles because they hinder most inventory enterprises. Usually, the involved taxonomists, because they are aware of their negative effects, spend a lot of time and energy trying to overcome them.

  • 2 Cf. Bogen 2013; Franklin 2002.

25There is still a different kind of general obstacle bearing on most inventories, that has nothing to do with practicalities, namely, theory-laddeness. We examine this kind in the remaining of this section. Theory-laddeness has been pointed out by many historians and philosophers of science, like Kuhn, Hanson, and Feyerabend, as having negative consequences on the relations between empirical data and hypotheses. Kuhn, Hanson, and Feyerabend have insisted that in order to take any observation as a piece of scientific data, it is necessary to entertain some hypothesis about the involved domain, otherwise it would be impossible to analyse the observation as contributing to the field of knowledge at hand. However, such an impregnation of data by hypothesis is potentially detrimental to the testing process. This problem has been widely handled by further philosophers of science who have carefully distinguished cases in which theory-laddeness has to be fought against from cases in which it is benign and does not prevent hypothesis-testing2.

26In the case of inventories of biodiversity, theory-laddeness clearly goes against the ideal of descriptive neutrality we mentioned in section 1.1. It is well-known that the ideal of descriptive neutrality is severely threatened by various sorts of biases and prejudices scientists may adopt against their own will, as well by the conceptual framework they have learned during their training years, whose validity may have been challenged by new data in-between. Psychologists have given powerful evidence about the influence of the way one conceptualises a situation on the way one perceives it. However, it may be argued that scientific training, especially training in taxonomy, aims to remove these various biases and to make biologists able to provide bias-free descriptions. Let us accept that training has this purifying effect for the sake of argument. Does it mean that taxonomists provide their biology colleagues with exclusively factual descriptions of species and the place they live in? We will discuss this question at length in the following section, in which we will focus on the less explicit and more hidden components of the WBK and on their influence on inventory practice.

27Before beginning this discussion, let us make clear that in this paper, we focus on the first step of any inventory, namely, collecting, as a way either to assess how many species live in a given area, or to further evaluate diversity, distribution, and abundance. We want to question whether collecting is as isolated from the WBK as it looks to be in the idealised notion of an inventory. We do not want to argue that taxonomist work is irremediably theory-laden, and thus unfaithful to what species are and where they live in. We do not have any general argument to this purpose. Our aim is different: we want to examine a particular example of an inventory task and disclose its specific difficulty. This example is the exploration of Pacific deep-sea fauna. Among the difficulties taxonomists trying to discover the species living in Pacific deep-water have to face, we find of course practical difficulties of access, but also obstacles of a more intellectual kind. We focus on the latter. These obstacles are rooted in more or less currently accepted assumptions, some of which are even not perceived as assumptions but rather as non problematic facts. They partake in the WBK and influence the way the inventory task is carried out. It is thus important to examine the nature of these assumptions and the epistemic attitudes taxonomists entertain toward them. This is our program for section 2.

2. The inventory of Pacific deep-sea fauna and the lasting influence of the azoic hypothesis

28Our aim in this section is to focus on an actual example of marine inventory in order to display various biases that do not enter into the general category of well-known obstacles, but remain unseen by the involved taxonomists and influence their work. Realising the role of these biases clearly conflicts with the common, idealised view that taking an inventory consists in describing brute facts. Moreover, these biases are difficult to detect for the reasons we present below, unlike the ones we have presented at the beginning of section 1.2. Our main example will be a set of hidden assumptions still bearing on inventories of deep-sea fauna even though the hypothesis they originate in, the “azoic hypothesis”, is known to be false. We begin by presenting the controversy over this hypothesis and go on with examples showing its contemporary influence when it is combined with economical constraints.

2.1. The controversy over the azoic hypothesis and how it went on shaping deep-sea research even after it was over

29According to the azoic hypothesis, proposed by Edward Forbes in the 1840s, deep sea is too hostile an environment to make any form of life possible. This hypothesis agreed with commonly accepted assumptions of the time, according to which it is impossible that organisms develop without light, as light was seen as an indispensable source of energy for organismal development. Moreover, weak temperatures and high pressures were also conceived as hindering the possibility of life.

  • 3 Forbes1844: 167.

30As a naturalist, Forbes was eager to develop what may be called an “ecological” approach even if the word was not yet used at the time. He was indeed interested in the distribution of organisms with respect to depth, by analogy with the distribution of organisms with respect to altitude. He was thus expecting that there are fewer organisms as depth increases, in the same way as there are fewer organisms as altitude increases. In order to test this hypothesis, he conducted an expedition on the HMS Beacon in 1841 in the Aegean. Here is his conclusion: “The number of species and of individuals diminishes as we descend […] pointing a zero in the distribution of animal life as yet unvisited”3. According to his computations, there was no life left under 600 meters; as a result he coined the name “azoic limit” to refer to depths below 600 meters.

31Today, we can easily understand why the data gathered by Forbes led him to hypothesize the existence of an “azoic limit”: as it happens, he had collected specimens in an especially nutrient-poor area. Moreover, the dredger he used was ineffective, for, on the one hand, it was small, and on the other hand, it did not allow him to collect even small organisms on soft grounds because it caught too much sediment and could not filter it correctly. For these reasons, he could not find living specimens under 600 meters.

  • 4 Anderson and Rice 2006.

32As Forbes was an influent member of the Royal Society, the azoic hypothesis was almost unanimously accepted within the scientific community even though specimens had already been discovered in deeper waters since the beginning of the nineteenth century4. However, as we shall see, it was difficult to assess the reliability of depth measurements, so that it was uncertain whether the specimens had been really caught below 600 meters.

33Another reason why Forbes’ hypothesis immediately became very popular is that it agreed with background knowledge and hypotheses about the conditions at which life is possible. However, he had some opponents who were convinced that the azoic hypothesis could not be true. Living organisms had indeed been collected of which it had been claimed that they lived in deeper water, in spite of the difficulty to evaluate under-water distances. Because of these lasting, material difficulties, the controversy could not be settled. But at the time, many marine biologists were aware of the issue and of the necessity to develop reliable means of evaluating depth.

34Let us illustrate how powerful the assumption of the impossibility of deep-sea life was. In 1818, Captain John Ross, exploring Baffin Bay, collected sea-stars and annelid worms at 1495 meters (according to his estimates). He used a device equipped with jaws allowing him rather accurate (though overestimated) depth measurements, as well as more efficient specimen catching. It has been found later that the specimens he collected lived at 1095 meters. However, it was so controversial to claim that organisms could live as deep as 1000 meters below the surface that the naturalists participating in this expedition decided not to publish their discoveries.

35Nevertheless, after 1844, the controversy quickly developed toward the failure of the azoic hypothesis. Just after the publication of Forbes’ paper, many naturalists collected specimens living in deep water. Between 1850 and 1868, Michael Sars from Oslo collected specimens from hundreds of different species living between 350 and 550 meters along the Norwegian costs. In 1860, George Charles Wallich collected 13 sea-stars at about 2305 meters. Here is his conclusion:

  • 5 Wallich 1862: 68.

This founding far exceeds in importance any previous sounding record. It proves the fallacy of several of the conditions which have heretofore been supposed to limit the bathymetrical distribution of animal life, and points to the existence of a new series of creatures peopling the deeper abysses of the ocean5.

  • 6 Milne-Edwards 1861: 153.

36At the same time, a telegraph wire connecting Sardinia to Algeria broke between 2000 and 2800 meters. The specimens which were living on the wire were analysed at Paris Museum by Alphonse Milne-Edwards. In his report, entitled Observations on the existence of diverse Molluscs and Zoophytes at very large depths in the Mediterranean sea, he mentioned both Forbes’, Ross’, and Wallich’ works and added that the analysis of the wire showed that “anthozoa from the scleractinian division, as well as molluscs, both gastropods and acephalous, are able to live in the Mediterranean below 2000 meters, and can even develop and grow at a fast rate”6. Milne-Edwards’ reputation was such that Forbes’ hypothesis began to falter.

37It is only at the end of the nineteenth century that the azoic hypothesis has been definitely refuted, due to the success of the Challenger expedition (1872-1976). The expedition began in the context of the controversy about the existence of deep sea life and the related question whether there are limits to life. Most naturalists expected that this expedition could refute Forbes’ azoic hypothesis, but it remained to understand how life could exist in such a supposedly hostile environment.

38The history of deep-sea fauna inventories nicely illustrates how some assumptions may frame the inventory task, which is however supposed to result in neutral descriptions. With the example of the azoic hypothesis, we have seen how assumptions, when taken for granted without further reflection, can distort the image of deep sea fauna emerging from the inventory investigations. What is striking in this example is that the rebuttals of the azoic hypothesis have been available for a long time. In this case, there is thus a tension between the erroneous representation some taxonomists develop as a result of an uncritical, and even partly unconscious acceptance of the azoic hypothesis and the representation they should promote in view of available evidence and current discussions in the relevant literature.

39The influence of the azoic hypothesis reveals a cognitive gap that plays an important role in the epistemic process actually implemented during deep sea inventories. The gap is between hypotheses that are potentially justifiable and that have at least a plausible foundation and assumptions that are uncritically accepted, like the azoic hypothesis. From this example, we can emphasise that it is usually difficult to clearly distinguish between the two sides of the gap, at least without hindsight.

2.2. Joint effects of economical constraints and the (remnants of the) azoic hypothesis

40In this section, we show how the influence of the azoic hypothesis has been so strong as to overcome the effect of the discovery of hydrothermal springs and their associated rich biodiversity in the late 1970s. Because hydrothermal springs house an unexpectedly rich fauna, it seems that the azoic hypothesis could not possibly be mentioned at all after 1977. However, it has been a main ingredient in a spectacular trial in Papua New Guinea as late as 2011. We present this affair in the following in order to illustrate the negative effects on taxonomic work of the survival of a refuted scientific hypothesis when combined with strong economic constraints.

  • 7 Desbruyères 1993.

41The discovery of the hydrothermal springs in 1977 revealed a surprisingly rich fauna living in extreme conditions: the hydrothermal fluid issuing onto the sea floor at these vent sites is hot (up to 390°C), anoxic, often very acidic, and enriched with hydrogen sulphide (H2S), methane (CH4), and various metals (especially iron, zinc, copper and manganese). It had an important impact on the way non-specialists thought about life in deep-sea. They realised that deep-sea was not the desert they thought it was (Laubier, 1984). The revealed fauna was both so spectacularly abundant that hydrothermal springs were called “oasis”. Thus, in the 1990s, there was no doubt left on the high diversity of deep-sea fauna, as testified by the following quote: “This biodiversity is comparable to that of tropical forest, but much richer in terms of biomass”7.

42Nevertheless, despite this discovery, the azoic hypothesis was still not dead, as we will now illustrate.

The Papua New Guinea case

  • 8 Tse 2007; Laznicka 2010.
  • 9 Report 1999. See Reichelt-Brushett 2012.

43Papua New Guinea is an example of a state where mineral-rich ores are abundant8. Consequently, the development of mine sites in this area is a potential source of income to local communities and regions. In 1999, the Papua New Guinea government lodged an application for a special mining lease for the Ramu Nico project, an important local company at that time. In 2000, the special mining lease was granted to Ramu Nickel Ltd. By this agreement, after a very controversial report9 made in 1999 by independent scientists from the Lutheran University of Papua New Guinea, the Papuan government approved the tailings disposal to the deep-sea ocean. The main argument in this report was that the effect of the mining lease was to be negligible because there is not much life in deep waters – a first, surprising revival of the azoic hypothesis, but not the last. The mine, the pipeline, the refinery, and the wharf were constructed from 2008 to 2010.

44In 2011, a trial was launched by a Papua New Guinea citizen against the Ramu Nico company to restrain from operating a deep-sea tailings placement (DSTP) system. Looking into the tribunal’s conclusions, we can see that the scientific report that was demanded in order to evaluate the impact of the release on deep-sea life drastically underestimated the importance of the benthos (deep) environment. The scientific report clearly stated that the benthos environment is less important, in terms of biodiversity, than shallower water environment. A consulting marine scientist, called as a witness by the defendants, declared that

  • 10 The last quote is from Ian Hargreaves, Consulting marine scientist, hydrologist, witness called by (...)

he does not believe that the DSTP will have any shallow water impacts – the environmental impact will be restricted to deep water: the smothering by tailings solids of the seafloor of Basamuk Canyon of approx 160 square km, which represents the ‘tailings footprint’ – benthic organisms, which already live in a very high natural sedimentation zone, may not be able to cope with an increased rate of sedimentation – as to the biological impact, DSTP is designed to protect the euphotic zone where 90% of marine animals live; however, no complete baseline surveys have been done of biological communities on the seafloor at Basamuk Canyon.
Though there will be to a large extent a smothering by the tailings of the sea floor at Basamuk Canyon not all benthic organisms will be destroyed, as many will be able to swim away, and the degree of biodiversity on the sea floor is not nearly as great as in the euphotic zone, which will be unaffected by the tailings, so there will be no significant adverse effect on the ecology of Astrolabe Bay.
Effect on the benthos: This does not represent a major problem, the defendants argue, as not all of the benthos will be eliminated (there will be organisms, including deep-sea fish, that will be able to move to areas unaffected by the tailings) and in any event the evidence suggests that the quantity and diversity of organisms in the deep-sea is quite low compared to the euphotic zone. Thus, while the ecology of the canyon may be altered, there is no evidence that it will be an adverse effect or if it is adverse that it will be substantially so10.

45This testimony clearly reveals that the azoic hypothesis was still very lively during this trial. For sure, it was not mentioned by taxonomists; however, the above-quoted trial-conclusions clearly indicate that the refutation of the azoic hypothesis has still not reached every scientist with a professional interest in deep-sea. As such, this fact has no consequence on the WBK as most biologists are conscious that deep-sea biodiversity is rich and abundant. Nevertheless, the fact that the refutation of the azoic hypothesis has not extended outside the biological community is likely to have important effects on their own scientific objects.

A new context for scientific research

46The Papua New Guinea case we have just presented nicely illustrates another, very important aspect of today’s deep-sea inventory practice, its being constrained by economic pressures. Not only does the Papua New Guinea case show how influent a refuted hypothesis can be, it is also typical of the way economic constraints can bear on inventories and of the intimate mixing between economic constraints and scientific requirements that currently shapes this practice. In order to analyse the precise effect of the entanglement between economic pressures and scientific life, we now turn to a last example. This example will provide us with an illustration of our last category of potential bias influencing deep-sea inventories.

47Facing the decrease of halieutic resources near the costs (since the 1960s), and considering the new mining prospects revealed by the discovery of the hydrothermal vents, the deep-sea environment has become important for new economics actors. As a result, the conditions of its scientific exploration have been deeply transformed. Besides their importance in terms of biodiversity, hydrothermal springs are also considered a rich reservoir of metals and their discovery has opened important mining prospects. The transformations introduced by these activities are clearly described in the following quote:

  • 11 Van Dover et al. 2014.

The deep-sea – defined here as ocean beyond the shelf break and depths greater than 200m – is increasingly recognised as a fertile area for offshore industrialisation. Current or future activities include fishing, waste disposal, cable lays associated with telecommunications, scientific research, oil and gas development, bioprospecting, mineral extraction, and tourism.11

48As for scientific research, the impact of this new situation is the following. The (maybe surprising) effect of the recent awareness of the biodiversity crisis has been to foster stronger links between scientific exploration and economical exploitation, issuing in entangled practices which are the new context in which deep-sea scientific investigation is carried out. For instance, it is more and more difficult to launch a deep-sea exploration program without the financial aid of a company. Conversely, economical exploitation increases the demand for scientific expertise based on exploration:

  • 12 Ramirez-Llodra et al. 2011.

The deep sea was (and still is) perceived as a service provider at two levels: (1) it served as a convenient site for disposal of waste, especially where land options were not politically and ‘‘ethically’’ attractive and (2) it was seen as a source of potential mineral and biological wealth over which there was no national jurisdiction12.

  • 13 Report 2013.

49This results in a highly paradoxical situation. Let us illustrate it by a French example. The first element of this example is that the Total and EDF (now VEOLIA) companies are strongly interested in the exploitation of deep-sea minerals. While the exploitation of metals in the hydrothermal sources is not rentable enough, the cobalt crust is a more interesting target. Found on seamounts, it seems easier to extract. In order to do so, it is necessary to receive the relevant agreements. A collective expertise (ESCO, in 2013)13, involving the CNRS and other institutions, has thus been demanded by the French government. It provides a synthetic report on currently available knowledge on deep-sea mineral resources and on the implications of their exploitation.

50An important conclusion of this report is that, as it is now commonly acknowledged that mineral exploitation has to care about the preservation of the biodiversity, a better knowledge of deep-sea life is required. As a result, the French companies are becoming important donors for deep-sean inventories. This is for them the best way to be able to develop deep-sea mineral exploitation. In turn, deep-sea biologists need this funding in order to take further deep-sea inventories. The paradox is that deep-sea biologists are both asking for money from the companies and responsible for the agreement about the companies’ deep-sea economical activities. A strong conflict is thus emerging, of which the French case is only one example among many others.

51Even though it is not entirely clear what the effects of the entanglement between economic exploitation and scientific exploration could be, this situation cannot but be seen as a source of potential bias on deep-sea inventories. Far from being uniquely interested in deep-sea fauna, taxonomists are required to be also interested in imagining the future actions of the involved companies.

3. Concluding remarks

52In this paper, we have presented examples showing some ways actual biological inventories can diverge from ideal ones. We have been careful to distinguish general constraints that bear equally on all biological inventories, like lack of time and of material resources, from constraints that only bear on specific cases. We have also introduced another distinction, between constraints that are explicitly taken into account by the researchers and constraints that are hardly recognised as participating in their background assumptions. Whereas the former can sometimes be partly overcome, the latter, whose influence is not acknowledged, are more difficult to counteract. In this last, short section, we try to draw some lessons from our above analysis.

53A major conclusion of our analysis is that the various pieces constituting the WBK are not all explicitly articulated each to the others. Even though taxonomic work is well-grounded on the theory of evolution on the one hand, and on the Linnaean nomenclature framework on the other, it is also more loosely articulated to other sets of hypotheses and constraints, some of which are not even identified as background assumptions. As an example, we have mentioned the azoic hypothesis. Its influence is partly hidden to the eyes of some of the researchers, who are not aware with its conflicting with other, more firmly grounded, hypotheses. When explicitly accepted hypotheses conflict with implicitly entertained ones, the capacity of inventories to provide basic knowledge for further biological knowledge is threatened.

54At the end of the paper, we have also put forward a further potential source of bias: economical constraints. These bear on the difficult articulation between the constitution of scientific knowledge and the demand for expertise in an emergency context, caused by the biodiversity crisis. Due to the growing awareness of the necessity of preserving biodiversity, scientific expertise is more and more searched for. However, scientific expertise on deep-sea biodiversity is doomed to rely on highly incomplete scientific knowledge. It thus requires more research to be done. As a result, the companies in which the demand for expertise originate are engaged in scientific research funding. This is undoubtedly the crucible of many epistemic conflicts.

55It seems to us that the best way to lessen the importance of the various biases we have identified is to first become aware of their existence. As we have emphasised at the beginning of section 1.3, there are some constraints, like the Ocean’s immensity, that bear on all inventories but whose influence is explicitly taken into account by the researchers. This does not mean that the constraints disappear, but that the taxonomists know that the resulting inventory is valid up to the limits imposed by the constraints. The awareness of these limits is included in the global knowledge taxonomists draw from the inventory. We thus emphasise that it is important to care about one’s background assumptions even in scientific tasks that look as basic as taking inventories. In other words, even working at the lowest level of the WBK, it is important being able to make every piece of the WBK explicit.

Torna su


Anderson, T.R. and Rice, T.
– 2006, Deserts on the sea floor: Edward Forbes and his azoic hypothesis for a lifeless deep ocean, “Endeavour”, 30(4): 131-137

Barberousse A. et Samadi S.
– 2014, La taxonomie et les collections d’histoire naturelle à l’heure de la sixième extinction, in Casetta, E. et Delord, J., La biodiversité en question. Enjeux philosophiques et scientifiques, Paris, Les Éditions Materiologiques: 155-180

Bogen, J.
– 2013, Theory and observation in science, in E.N. Zalta (ed.), The Stanford Encyclopedia of Philosophy (Summer 2014 Edition),

Carvalho, M.R. et al.
– 2013, Does counting species count as taxonomy? On misrepresenting systematics, yet again, “Cladistics”, 30(3): 322-329

Conclusions of the Radu Nico trial
– 2011, Papua New Guinea National Court of Justice

Desbruyères, D.
–1993, La vie au fond des Océans, “Pour la Science”, n. 192

Emig, C.C.
– 2010, De la renommée à l’oubli: l’effet baudruche,

Fontaine, B., Perrard, A. and Bouchet, P.
– 2012, 21 Years of shelf life between discovery and description of new species, “Current Biology”, 22(22): R943–R944

Forbes, E.
– 1844, Report on the mollusca and radiata of the Aegean Sea, and on their distribution, considered as bearing on geology, “Report of the British Association for the Avancement of Science for 1843”: 129-193

Frankin, A.
– 2002, Selectivity and Discord: Two Problems of Experiment, Pittsburgh, University of Pittsburgh Press

Laubier L. et Desbruyères D.
– 1984, Les oasis du fond des océans, “La Recherche”, n. 161

Laznicka, P.
– 2010, Giant Metallic Deposits: Future Sources of Industrial Metals, 2nd ed., Berlin, Springer

Milne-Edwards, A.
– 1861, Observations sur l’existence de divers Mollusques et Zoophytes à de très grandes profondeurs dans la mer méditerranée, lues à l’Académie des sciences le 15 Juillet 1861: 149-157

Ramirez-Llodra, E. et al.
– 2011, Man and the last great wilderness: Human impact on the deep sea, “PLoS One”, 6(8): e22588

Report 2013
– 2013, Expertise scientifique collective sur les impacts environnementaux de l’exploration et de l’exploitation des ressources minérales profondes,
URL =,38968.html [accessed November 2014]

Report 1999
– 1999, Environmental risks associated with submarine tailings discharge in Astrolabe Bay, Madang Province, Papua New Guinea. A discussion paper commissioned by the Mineral Policy Institute, Sidney, Australia

Tse, P.-K.
– 2007, The mineral industry of Indonesia. Pp. 13.1-13.10, in US Geological Survey 2005 Minerals Yearbook – 2005,
URL = [accessed May 2012]

Van Dover, C.L. et al.
– 2014, Ecological restoration in the deep sea: Desiderata, “Marine Policy”, 44: 98-106

Wallich, G.C.
– 1862, The North Atlantic Sea-Bed: Comprising a Diary of the Voyage on Board HMS Bulldog, in 1860; Observations on the Presence of animal Life, and the Formation and Nature of Organic Deposits at Great Depths in the Ocean, London, John Van Voorst

Torna su


1 Cf. Carvalho et al. 2013.

2 Cf. Bogen 2013; Franklin 2002.

3 Forbes1844: 167.

4 Anderson and Rice 2006.

5 Wallich 1862: 68.

6 Milne-Edwards 1861: 153.

7 Desbruyères 1993.

8 Tse 2007; Laznicka 2010.

9 Report 1999. See Reichelt-Brushett 2012.

10 The last quote is from Ian Hargreaves, Consulting marine scientist, hydrologist, witness called by the defendants; Conclusions of the Ramu Nico trial.

11 Van Dover et al. 2014.

12 Ramirez-Llodra et al. 2011.

13 Report 2013.

Torna su

Per citare questo articolo

Notizia bibliografica

Anouk Barberousse e Sophie Bary, «Ideal and actual inventories of biodiversity»Rivista di estetica, 59 | 2015, 14-31.

Notizia bibliografica digitale

Anouk Barberousse e Sophie Bary, «Ideal and actual inventories of biodiversity»Rivista di estetica [Online], 59 | 2015, online dal 01 août 2015, consultato il 18 juin 2024. URL:; DOI:

Torna su

Diritti d’autore


Solamente il testo è utilizzabile con licenza CC BY-NC-ND 4.0. Salvo diversa indicazione, per tutti agli altri elementi (illustrazioni, allegati importati) la copia non è autorizzata ("Tutti i diritti riservati").

Torna su
Cerca su OpenEdition Search

Sarai reindirizzato su OpenEdition Search