| Human Nature Review ISSN 1476-1084 | Table of Contents | What's New | Search | Feedback | Daily News | Submit A Manuscript |
PDF of this article
Download Adobe Acrobat Reader
Email the reviewer
Reviewer's web site
Search for papers by Gold, A. M.
Author's web site
Search for papers by Hey, J.
Publisher's web site
Send a response to this article
Search the web for related items
Contact the Editors

The Human Nature Review  2002 Volume 2: 135-143 ( 10 April )
URL of this document http://human-nature.com/nibbs/02/gold.html

Essay Review

Epistemology among the Lumpers and Splitters


Bert Gold*, Ph.D., F.A.C.M.G.

Genes, Categories, and Species:
The Evolutionary and Cognitive Causes of the Species Problem

By Jody Hey
217 pp, Oxford University Press (2001)

In 193 pages, Jody Hey, a Professor of Genetics at Rutgers University, takes us on a tour of the philosophy, science, and some of the mathematics of species and speciation. During his tour, Hey provides us with lucid moments of pyrotechnic effect that crackle with vivid imagery. However, the essential thesis, that ‘the species problem’ is a consequence of linguistic confusion, a blurring of the distinction between holophrastic language and sense, appears less than proved to this reader.

John Maynard Smith’s introduction endorses the work and expresses understanding of Hey’s focus. Maynard Smith discusses his own dissatisfaction with species definition while completing a series of collaborative experiments on Neisseria, some time ago. Hey begins with the question of dog species definition and asks whether copulatory incompatibility is sufficient: He uses the example of an oversized Mastiff paired with a diminutive Chihuahua.

Perhaps the quickest encapsulation of the ideas of the book appears near the end, when Hey is discussing the difficulties that the species problem imposes on systematics.

If there is a species of tree in the forest, and no systematist is around to diagnose it, is it still a species?

In this summing up, Hey discerns that the existential problems of the systematist undermine practical application of genetics and ecology to conservation biology. So, instead of solving the ”species” problem in a practical way, internally, through reference to genetics, morphology, homology, and phylogeny, Hey resorts to epistemology and linguistics for a redefinition of the relevant issues. What he gains is redefinition, but I am not certain that in practical effect this provides much general utility.

It is rather like what Evelyn Fox Keller gave us in her Century of the Gene (1). Fox Keller implores us to examine the shifting definitions of the gene throughout the century and throughout her work. In a way, her case may be stronger than Hey’s, as she presents convincing arguments that Johannson’s coining the term ’gene’ in 1909, and the arguments that followed on about it, continually provided redefinition. She suggests that genetics, having reached its apotheosis in the completion of the Human Genome Project, has stretched the definition of the word ‘gene’ beyond its breaking point. For Hey, the term ‘species’ carries with it a similar level of fuzzy nuance. And yet, at the end, Hey has provided no more than Fox Keller has: A discourse on design of the vocabulary of discovery, with all its imprecision stripped bare for everyone’s gaze. In a way, all rigorous science is subject to the same vagaries of linguistics and nuance of language as Hey has deconstructed here.

The book is in three parts. During part I, Hey introduces ‘The Hidden Question’ which derives from the word “species” itself. We are reminded of the word’s etymology, from Roman translation of the Greek, eidos, used by Aristotle and Plato to refer to categories of things. Hey recalls Linnaeus’ and Darwin’s efforts to define species and frames the issue as a kind of workaday problem for naturalists, i.e.- in practical use ‘How can I fit this new specimen from my collection into the existing schema?’ The Biological Species Concept (BSC) of Mayr (2) is invoked, as are many other extant methods boundaries between species. It is not until chapter 2 in his discussion of measurement that Hey begins to come to grips with the problem. For, as John McCarthy has so eloquently stated: ‘He who refuses to do arithmetic is doomed to talk nonsense’ (3). Hey begins with Popper’s 1959 observation that applies John McCarthy’s comment differently: “Whenever measurement is undertaken, there is a hypothesis being tested”. Here, almost all the hypotheses concern dividing lines between the intangible and the uncertain. Measurements are too often efforts to create boundaries where none apparently exist. Hey deals in general and in specific terms with this topic, even discussing count creep, the tendency for biologists to look at new data and define a heretofore singular as actually more than one entity. It is this tendency that gives rise to the term ‘splitter’ in the title of this essay.

Hey tells us about his own work on Drosophila and confides that, although his genetic experiments have often found evidence of disparateness in anecdotally described ‘species’; that occasionally he finds no such evidence. He avers that this does not disturb him, as he expects these inconsistencies to better define what a ‘species’ is (p. 5). But, Hey chooses not to discuss the oft-cited example of sympatric speciation in the Hawaiian Islands extensively studied by Hampton Carson (4). As confounding observations often motivate progress in science, I was surprised that Hey was not more forthcoming with those Drosophila species observations that he has found discrepant.

My own experience with the ‘species problem’ dilemma stems from two ostensible species of Ranunculus, with leaf morphology unique to their habitation, that I studied as a graduate student. One of these grows on woody highlands and one on marshy lowlands. In this case, phenotypic plasticity and not genetic uniqueness confer disparate leaf morphology. Transplantation experiments provide a method of differentiating species from subtypes (5). Two forms, one species, one genetics. But, many times in history these were split in two: It took long to realize that the terrestrial and aquatic forms were each of the same stock. In fact, Stebbins (6) provides Ranunculus to us as an example of a species with blurred borders: one that cannot easily be classified, no matter our level of mathematic rigor. And then, refreshingly, Hey tells us about Stebbins (7), using the brilliant analogy conceived to illustrate why all Biology was in such a muddle over species: While Gregory Bateson (8) was writing his celebrated ‘metalogues’, such as ‘Why does everything always get in a muddle?’ and ‘Why a Swan?’ Stebbins was writing an allegory to help biologists understand the problem of species and speciation. Stebbins wrote that we should imagine a bright day with a clear blue sky and well-defined cumulus clouds. Surely, we would have no problem discerning the borders then. But, should we fly up close into them, the borders, though distinct, would be less obvious, and ultimately, they would formulate as regional differences in water vapor density. Back down on earth, a view of a cloudy day might make even the most rigorous meteorologist despair of ever doing cloud arithmetic again; as a series of indistinct clouds are not countable, do not border on other clouds, each one melding together as so many species in the wild. It is from this analogous view, that Hey begins to approach ‘the species problem’. After providing Stebbins’ allegory, Hey’s pace noticeably quickens. He provides some thoughts from Quine on language and logic (9), mention of Crick’s Astonishing Hypothesis (10) (which Minsky speaks of as ‘The Vision Book’), also some rudimentary ideas about molecular genetics plus a brief discussion of Dawkins thought on what is required it takes to make a unit which survives. Hey ends the first part of the work with a shocking comment, one that he will return to in later chapters with emphasis: Because there is no one generally accepted species concept, biologists have been struggling. And, like some half-formed definitions of pornography, there are biology adherents of the ‘I know it when I see it’ school (11). In an effort to provide definition, Mayden (12) has catalogued some 24 different concepts of what a species is, and yet, that list is clearly incomplete according to de Queiroz (13). Hey ends this section stating that we have far to go on our journey to learn and define a species.

Hey begins the second part by reminding us of the fundamental qualities of language. Without rebuilding our framework from Mach (Ding an sich) (14) to the Whorfian hypothesis (15), and without indulging in detailed discussion of either Chomsky or Piaget (16), Hey leaps to Quine’s simplification in Word and Object (9), asserting that language need not be perfect in order to use it in incremental ways to describe the world. Then, as if borrowing a phrase from Nicholas Negroponte’s Being Digital (17), Hey begins to describe the fuzzy logic inherent in our bytes:

…language, whether spoken, signed or written, is essentially digital. It is made up of a finite set of discrete components, symbols or sounds, and it carries information primarily in the sequence of those components. Since language is discrete, and much of the real world is continuous, there are necessarily a great many circumstances where language can only be an approximation.


Could the mismatch between discrete language and fuzzy reality be the cause of our species problem.

Hey ends the latter sentence with a period, just as I have done. He is not actually asking the question, merely arguing that we are mistaken to think that we understand the species problem to be one resolvable through morphology or anatomy. Hey is alleging that the study of homologous organs or molecular analysis is, in a sense, passé. He urges us to recognize that the limits of our species definition derive from our imprecise language and that fuzzy thinking will not easily be overcome.

Much of the second part of the book has to do with categorization: How we create them, how universal they become and why they make sense. Hey cites Jared Diamond’s work among the Fore people in New Guinea (18), where it was discovered that the vast array of native plants observed and named by the aboriginal settlers almost overlapped completely with the plants identified by European ‘experts’. Most disinterested observers might conclude that either the Europeans were conveying hidden messages to the aborigines, or that there is a kind of taxonomic Universal Grammar (UG) of the sort that Chomsky postulated to explain man’s capacity to learn language and understand ideas. (Chomsky never suggests that we know words ‘in utero’ or that language itself passes through the genes, but rather that some formidable apparatus exists which makes us eligible for learning and thinking). Hey skips this opportunity for detail (however, Diamond, in his works on evolutionary subjects, discusses it at length), instead preferring to focus upon the non-arbitrary nature of name assigning systems and the hierarchical arrangement found in most taxonomies.

The crux of the species problem, that of tree building, of fractals, chaos theory, and the theories of speciation occupies Chapter 6, entitled ‘Biological Diversity’. Hey chooses not to derive rigorously the coalescent, instead preferring to keep his book popular rather than solely for the specialist. The focus of discussion remains reframing the species problem as a linguistic conundrum. The sixth chapter begins telling us:

The geneticist Alan Templeton once defined species as ‘the most inclusive population of organisms having the potential for phenotypic cohesion through intrinsic cohesion mechanisms’.

Hey decries this popularly acclaimed definition. He unmasks it as a tautology framed in the ostensibly palpable, yet suffering a lack of utility. Hey urges us to substitute the word “cohesion” with any other word and suggests we will find equal meaning. (Ghiselin later made efforts to repair Templeton’s mis-construction by identifying a species as that which is by what it does, specifically defining species as the products of speciation.) Hey reminds us that both Ghiselin and Templeton are insistent that species are real (and not only linguistic) entities, yet:

Their method could be a page from the tautology cookbook: Take one strong species presumption and an equal measure of intangible reality, and squeeze out a definition.

Hey then explains modern day phylogenetic tree building. Hey’s teaching skills are very strong. In the remainder of the chapter, he makes efforts to represent species as evolutionary groups, deriving from homology comparisons. With an interesting narrative and several illustrative diagrams, Hey informs the reader of how to use the neutral theory and genetic drift to define organismal divergence times. He reminds us time and again that speciation is the property of the population and not of individual organisms. He places each classified organism in the context of an evolutionary hierarchy. During discussions of his area of expertise, Hey becomes truly illuminating; as an expert in recombination genetics of the kind that contribute to the rise of species, he presents several compelling arguments and explanations for the evolution of individual chromosomes, the necessity of recombination as a pacemaker for evolution and the central import of geographic, temporal or reproductive isolation in evolving populations. Hey discusses Wright’s isolation by distance model to illustrate the importance of boundaries in defining evolving groups. Thereby, he makes certain to better define for us the shift between a sharp partition and one where drift, competition and adaptation may make differentiating groups more difficult. Boundaries become the essence of the issue for Hey, as they define the difference between the clouds we can plainly discern, and being surrounded by Stebbins’ fog. In Hey’s view the world of language as well as our ability to enumerate biological difference as a consequence of ecologic niche, provides an underpinning for species identification and speciation.

Not all biologists are uncomfortable with real imperfect boundaries between kinds of organisms. One of the main reasons for this is the existence of theories on the gradual origin of the species. Any theory where one thing gradually becomes two, necessarily admits the existence of real partial boundaries, of circumstances where counts cannot be made.

Hey turns to Venn Diagrams, Fibonacci series, and fractals to remind us that although we have reached a sort of sophistication in our classification schemes, our progress may be illusory. Describing the Scottish landscape and defining the catalog of peaks over 3000 feet high established by Sir Hugh Munro in 1891, Hey reminds us that the current day accounting of 284 such Scottish Peaks is itself a simplification: The movement of some rocks at decisive landscape levels in the Scottish highlands could radically alter our perception and, as a consequence, the numerology of Munro peaks counted. Summing up this discussion, Hey tells us that:

If biodiversity is fractal, then we would expect that one could always find a finer pattern within a pattern, a smaller group within a group, except in the not very useful limit wherein basal taxa include just individual organisms.

In his 7th Chapter, Hey more rigorously defines the problem of blurred species and elucidates the meaning of recombination for speciation. He calls this chapter ‘Recombination and Biological Species’. This material is evidently at the core of Hey’s scholarship. Without semantic arguments about whether the germ cells are somatic parasites, or whether the soma is an empty vessel intended for conveying germ cells (Bateson), Hey reminds the reader that selection occurs in at-risk populations as a function of the level of genetic novelty present:

Otto and Barton (1997) modeled the evolutionary fate of a mutation that is a neutral modifier of recombination and that occurs in a finite population of chromosomes that are segregating multiple selected mutations. They showed that a mutation that has no other effect than to increase recombination will have an increased chance of fixation, relative to a neutral mutation that does not alter recombination.

Hey is here restating the old observation that recombination has survival value for the species at risk. Sex is important for survival. Yet, Hey makes the observation rigorous, and does it with panache. By discussing recombination in this light, Hey avoids harping on the tendency of recombination to unproductively homogenize the genome in a reproducing population. He emphasizes difference and speciation. This is his strength. That he can provide a rapid view of the mechanics of the process in order to emphasize the problem of defining species. In a subchapter on “The effect of recombination on the boundaries of evolutionary groups”, he reverses his emphasis on difference and returns to the theme familiar to most biologists in the context of recombination and homogenization. Hey’s analyses become particularly erudite; his writing remains complex, but he argues the issue of speciation, much like the population genetics master, Sewall Wright. Hey reminds us that:

…two populations cannot appreciably diverge if they exchange genomes at a rate much higher than the average of one per generation.

By citing Wright’s work and of the problems of defining a recombining population, Hey moves us forward by recalling the work of the Grants on Darwin’s Finches in the Galapagos, and of their students, such as Dolph Schluter on Stickleback fishes. The sum of this evolutionary observation and experimentation establishes the importance of niche definition, environmental change, reproductive isolation, and hybrid disadvantage for successful maintenance of speciation. Throughout this presentation there is the voice of Dobzhansky, advocating for the species and organism by proposing that reproductive groups are the natural basis for the species concept. Hey reminds us that Dobzhansky and Mayr were partners in the development of the Biological Species Concept; Dobzhansky focusing upon the genetics which drive speciation and Mayr concentrating more on reproductive isolation through temporal, biogeographic, and ecological sequestration.

The remainder of the book largely consists of an interior monologue between Evolutionary Group Theorists and those who advocate the Biological Species Concept. Mayr’s ideas are so central to this section, it is surprising that Hey postpones them until page 102. Hey reminds us that the ideas concerning the Biological Species Concept have developed since its first introduction in 1942, but that each definition bears a commonality:

…a species is a group of organisms that reproduce with one another and that are not capable of reproduction with the members of other such groups.

With this insight, Hey has strongly focused this book. Nothing that he states in terms of the linguistic explanation for the species problem is likely to bridge the gap of knowledge about specificity of definition that species require beyond Mayr and Dobzhansky’s great generalizations. But Hey insistently returns to his linguistic digression:

The purpose of this book is not to define SPECIES1, but rather to explain why we have difficulty with the word.

This is just as he is about to introduce the term, “Evolutionary Group” - a cleaner solution to the difficult problem of using the word ‘species’ in all the shades of meaning intended. He continually alleges that the BSC falls far short of the goal of unequivocal species identification, stating:

A major concern is that the BSC does not serve for organisms that never or rarely engage in sexual reproduction or exchange genes with other organisms. But even if we ignore that major shortcoming, there remains an enormous practical difficulty. Most times when biologists would examine organisms in order to make decisions about species, it is (… a practical impossibility, sic.) to make assessments of whether or not they do, or can, share in sexual reproduction or recombination.

Hey is in part a ‘tree builder’, and so his discussions on Phylogeny and Systematics, which begin the third and final portion of his book, are a joy to read. He develops the theory of evolutionary groups to immense effect, and contrasts these with less meaningful Phylogenetic Systematics (PS). Ultimately he deconstructs PS on grounds that its practitioners engage in tautological removal from species reality, further elaborating:

With a misleading credo tucked close to the breast, PS practitioners go forth to solve the species problem by finding just the right words for definition of SPECIES that causes the fewest problems.

His chapter on systematics ends deep in epistemology. In a subchapter which he refers to as “The Concept Problem”, Hey begins to look at how to frame the species problem in a way that will make sense for the real world. One can almost see him struggling with the real implications of not defining species adequately -- that conservation and real knowledge about biological diversity will suffer. Yet, with a great deal at stake, he concludes:

What we have missed is an appreciation of our own role in devising categories, and of our own desires to have those categories be the entities in our theories. Evolutionary groups are just one major cause of our species taxa, and we are the other.

In the above statement Hey disempowers all those who would use the species definition to protect. Hey gives us his abridgement of the endangered species act, viz.:

There is one route that would preserve SPECIES for real entities and that would not equate it with all evolutionary groups, and that is simply to reserve it for large evolutionary groups. That is what I think we should do.

Yes, but what of the assertion that such large groups are composed of many smaller, less populous members? Hey commits himself to being a “lumper” with the following:

In response to the concern that the idea of big evolutionary groups seems too vague, it must be said again that we cannot presume more distinction than actually occurs.

An effort is made in the final chapter of the book, ‘What is to be Done?’ to provide a sense of optimism, to propose that the solution lies in looking at higher taxa, more difference, for resolution to the species question. Hey alleges that we cannot count species because there is nothing defined that corresponds well enough to what is present that it can be counted. He reminds us of Stebbins’ cloud analogy and suggests that counting species is a futile as counting clouds. But he ultimately provides some flexibility, for he is aware that the implications of the epistemologic conjecture reach far. Nearing the end of his lament, Hey tells us that it is reasonable to accept that “…a species taxon is what systematists agree it is…”. Here is an effort to recant: Realizing that his assertions on the arbitrariness of species assignments and the construction of the problem as linguistic or epistemologic may invalidate the practice of conservation, Hey attempts an explanation:

There are other ways to assess and to preserve biological diversity that do not presume knowledge of real species or of unambiguous species taxa. One way is to rely on the hierarchical taxonomic system and to work to preserve higher taxa that are more inclusive than species taxa. The case can even be made that, by taking care of more inclusive taxa, many of the species taxa that would be categorized within those taxa will also be taken care of.

Hey has thus given a survey of the term SPECIES, replete with epistemology that would satiate even the most ardent student of philosophy. He has also provided a deeper understanding of species to this reader, unfortunately it is one that may cleanse it of some of its desired effect upon conservation. As my devout Christian mathematics Professor deemed about Chance and Necessity (19), almost 30 years ago, a book may be judged by the moral system it implies. The morals of this book circumscribe the limits of language, the theoretical mechanics of speciation, the relevance of conservation, and perhaps the legal underpinnings of the Endangered Species Act.


1. I myself should engage in a rather short digression. In his preface, Hey makes a large distinction between the word species (lower case) and SPECIES (all small capital letters). He reiterates this will be tedious for us, his readers, but wants to make a distinction between the word species as symbol, and the word itself, with its natural referent. Therefore, he warns us, we should be on the lookout for small capital letters, for they make meaning of their symbols, and not to the referent.


1. Keller, E. F. (2000) The Century of the Gene (Cambridge, MA and London, UK: Harvard University Press). Amazon US | UK

2. Mayr E. (1957) The Species Problem: A Symposium Presented at the Atlanta Meeting of the AAAS, Dec. 28-29, 1955 (Washington, DC: American Association for the Advancement of Science) and Mayr, E. (1982) The growth of biological thought. (Cambridge, MA: Harvard University Press).  Amazon US | UK

3. http://www-formal.stanford.edu/jmc/progress

4. Carson, H. L. (1982) Evolution of Drosophila on the newer Hawaiian volcanoes. Heredity 48: 3-25.

5. Hutchinson, G. E. (1975) A Treatise on Limnology, Volume 3 (New York: John Wiley & Sons) pp 175-187.

6. Stebbins, G. L. (1974) Flowering Plants, Evolution above the species level (Cambridge, MA: Harvard University Press), p 44.

7. Stebbins, G. L. (1969) Comments on the search for a “perfect system”. Taxon 18: 357-359.

8. Bateson, G. (1972) Steps to an Ecology of Mind (New York: Balantine) Amazon US | UK

9. Quine, W. V. O. (1960) Word & Object (New York: John Wiley) Amazon US | UK 

10. Crick, F. (1994) The Astonishing Hypothesis: The Scientific Search for the Soul. (New York: Scribner). Amazon US | UK

11. Mr. Justice Potter Stewart in Jacobellis v. Ohio, 378 US 184 (1964).

12. Mayden, R. L. (1997) A hierarchy of species concepts: The denouement in the saga of the species problem. Pp. 381-424 in Species: The Units of Biodiversity. Edited by M. F. Claridge, H. A. Dawah, and M. R. Wilson. (London, UK: Chapman & Hall).  Amazon US | UK 

13. de Queiroz, K. (1999) The general lineage concept of species and the defining properties of the species category. Pp. 49-89 in Species. Edited by R. A. Wilson (Cambridge, MA: MIT Press). Amazon US | UK 

14. Mach, E. (1883) Le développement de la mecanique, Une traduction exist sous le titre: Le mecanique, (PARIS, FR, Hermann). P 454.

15. Whorf, B. L. (1956) Language, Thought and Reality, J. B. Carroll ed., (Cambridge, MA: MIT Press). Amazon US | UK 

16. Piattelli-Palmarini, M. ed. (1980) Language and Learning: The Debate between Jean Piaget and Noam Chomsky. (Cambridge, MA: Harvard University Press). Amazon US | UK 

17. Negroponte, N (1996) Being Digital (New York, Vintage) Amazon US | UK 

18. Diamond, J. (1997) Guns, Germs, and Steel: The Fates of Human Societies (New York: WW Norton & Company) Amazon US | UK 

19. Monod, J. (1972) Chance and Necessity. A Wainhouse, trans. (New York: Random House). 


The author wishes to thank Dr. Jill Pecon Slattery for helpful comments and the staff of the MBL/WHOI Library in Woods Hole, Massachusetts for generously extending privileges. Ms. Doris B. Gold provided editorial assistance. The content of this publication does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products or organizations imply endorsement by the U. S. Government. This project has been funded in whole or in part with Federal Funds from the National Cancer Institute, National Institutes of Health.

* Address correspondence to Dr. Bert Gold, Laboratory of Genomic Diversity, The National Cancer Institute at Frederick, Building 560, Room 11-85, Frederick, MD 21702, USA. Email: goldb@ncifcrf.gov

Buy Genes, Categories, and Species from Amazon United States of America Amazon.com  Amazon United Kingdom Amazon.co.uk  Amazon France Amazon.fr  Amazon Deutschland Amazon.de  Amazon Japan Amazon.co.jp Amazon Canada Amazon.ca

Computer-generated translation of this page French français German deutsch Spanish español Portuguese português Italian italiano Russian Russian JapaneseJapanese Chinese (Traditional) Chinese (Traditional)Arabic Arabic― also try this alternative fast translation service.

© Bert Gold.


Gold, A. M. (2002). Epistemology among the Lumpers and Splitters. Human Nature Review. 2: 135-143.

US -

Amazon.com logo

UK -

Amazon.co.uk logo

The Human Nature Review