The Definition of 'Kinds'

Issues associated with the creationist definition of Biblical 'kinds'.

Introduction to the Concept of 'Kinds'

One of the primary creationist arguments against the theory of evolution revolves around the postulation of discrete categories of organisms, known variously as kinds (from the term used in Genesis 1) or baramins (from the Hebrew word "bara", which means "created", and "min", which is generally translated as "kinds"). According to creationists all extant organisms are derived from a small number of originally created organisms, and evolutionary change is limited to 'variation within kind'. Thus a fly remains a fly and a mouse remains a mouse regardless of how long they are exposed to random mutation and selection.

Searching for a precise creationist definition of the term 'kinds' can be a frustrating affair. While there seems to be general agreement that kinds represent discrete categories of organisms that are somehow prevented from evolving beyond certain boundaries, it is difficult to obtain a definition that represents a genuinely testable scientific concept.

The True Origins definition of 'kind' is given as follows:

          kind  n.

the created “kind” (from the Hebrew word baramin) refers to the originally created populations of various forms of life from which all other forms have arisen. It does not deny variation or mutation, but says that instead of one unicellular organism being the proginator of all life on earth through all time, there were a number of originally created populations whose individuals cannot vary or speciate across the discontinuities which separate each kind from every other kind. The concept of baramin is related to the concept of discontinuities that exist between groups of organisms. For instance, the dog, the wolf, the coyote, are clearly in the same baramin. And there is a definite discontinuity between this baramin and the bovine baramin, although both are mammals. Baramins can be partially identified by successful (live birth) hybrids, but probably go way beyond what hybridization can do today. Genetic studies may help determine discontinuities. The fossil record is also a help.

This definition provides a very basic framework for research into a potential scientific definition for 'kind', but it is far too general to be of any use by itself. It certainly does not provide a definition that could be used to generate a robust and testable description of the delineation between kinds, and could not alone be used to definitively distinguish whether two organisms belong to the same kind.

Creationists who have commented on possible testable definitions of kinds, including Don Batten and Jonathan Sarfati from Answers in Genesis, generally advocate a definition of kinds based on such hybridisation studies. This seems to me to be a thoroughly unsatisfactory definition for a number of reasons.

• It requires extensive testing to be validated - hundreds or thousands of species would have to be subjected to hybridisation studies to delineate even a few of the basic kinds.

• It is unreliable - hybridisation studies are prone to false negative results.

• It is of limited applicability - this method cannot be used on asexually reproducing organisms, including most prokaryotes and many plant species.

It also possesses the disadvantage of being unfalsifiable - if two organisms thought to be members of the same kind are shown to be incapable of interbreeding it still does not exclude them from being in the same kind, according to Don Batten (from this site):

If two animals or two plants can hybridize (at least enough to produce a truly fertilized egg), then they must belong to (i.e. have descended from) the same original created kind. If the hybridizing species are from different genera in a family, it suggests that the whole family might have come from the one created kind. If the genera are in different families within an order, it suggests that maybe the whole order may have derived from the original created kind.

On the other hand, if two species will not hybridize, it does not necessarily prove that they are not originally from the same kind. We all know of couples who cannot have children, but this does not mean they are separate species!

Creationists might claim that precisely the same criticisms apply to the biological definition of species, which is widely used by biological scientists. Isn't it hypocritical to demand that a creationist term be defined so precisely when a term used by many evolutionists is just as vague and poorly defined? The answer is a resounding no, for a very simple reason - evolutionists require that the boundaries between species be loose and ill-defined, as without these boundaries being blurred there is no way for evolution to take place. The fact that taxonomic categories are generally poorly defined is actually evidence that the evolutionary paradigm is correct. In contrast, creationists insist that the boundaries between kinds are concrete and impassable and hence should be able to define both the borders of these discrete categories and the mechanisms that maintain them.

Nonetheless, creationists have performed some basic research into the viability of a definition of the originally created kinds based on hybridisation studies. Some of the most thorough work has been performed by the German Siegfried Scherer and published in his 1993 book Typen des Lebens ('Types of Life'). A review of Scherer's book by the creationist organisation Geoscience Research Institute is available online here.

Scherer claims to have used the results of hybridisation studies to determine nineteen 'basic types' of organisms which presumably correspond to discrete kinds. The nineteen basic types are listed in the table below (an asterisk indicates that the types are tentatively characterised, while the others are claimed to be "based on more solid evidence").

The exact techniques used by Scherer to obtain these examples of kinds are explained only vaguely, and in future creationists will need to derive exacting and precise methodologies for delineating the boundaries between kinds.

An Unpleasant Dilemma for Creationists

It is interesting to note that the definition of 'kind' as a group of organisms capable of interbreeding places creationists in a rather unpleasant dilemma. The essay 'Apes, Apemen and Men' by creationist Lee Spencer discusses the fossil evidence for the hominid lineage and concludes that:

The fossil hominid crania show a gradation of morphologies between those characteristics seen in the apes but not in humans, to those characteristics seen in humans but not in apes. There is also a gradation in endocranial volume. Similarly , the postcranial skeletons of the hominids exhibit many characteristics that are intermediate, as well as a mixture of other characteristics that are either ape-like or human-like. Therefore, it is difficult to maintain the position that there are no morphological intermediates between apes and man.

Spencer is obviously unwilling to conclude that these results are due to an evolutionary sequence between ape-like ancestors and modern humans, and is thus forced to adopt a rather unfortunate position to avoid this possibility:

The only explanation that is consistent with the data and with the Biblical account of origins, is the outrageous proposal that man has been engaged in the sordid enterprise of exchanging genetic material with the pongids, either in nature or in the laboratory.

Of course, in doing so Spencer commits a grave error. By claiming that humans and pongids (anthropoid apes) have interbred to produce fertile offspring, Spencer has satisfied the criteria for placing humans and pongids in the same kind! Creationists are thus torn between claiming interbreeding to explain the existence of seeming transitionals in the hominid fossil record, and needing to avoid this explanation as it places humans and pongids in the same baramin. This is a dilemma that is generally ignored by creationists rather than being addressed in any substantive way.

A Phylogenetic Definition of 'Kinds'

For a number of reasons described above the hybridisation technique of delineating kinds is an unsatisfactory one for both creationists and evolutionists. Certainly a more precise and testable definition is required in order for the notion of 'kinds' to become a genuine scientific concept with more general applicability. In the absence of any useful creationist information on such a definition I have decided to attempt to derive one of my own, using the principles of phylogeny.

Phylogeny is the study of the patterns of similarity among living organisms. Phylogenetic analysis uses features of organisms (such as morphological structures or molecular sequences) to construct a tree which depicts the similarities between the organisms being examined. More on the techniques used in phylogeny can be found here and here, while thorough depictions of the phylogenetic trees constructed for all living organisms can be found at the Tree of Life and UCMP sites.

It is widely accepted that phylogenetic analysis can be used to determine the pattern of relationships between extant organisms - organisms that are similar with respect to many independent features are thought to be closely related, while organisms that are dissimilar with respect to most of the features examined are thought to be only distantly related. In this way the comparison of the morphological features, and more recently the molecular sequences, of many different organisms has allowed scientists to construct a phylogenetic tree incorporating all extant organisms.

It seemed to me that there was a distinct possibility that kinds could be defined with reference to phylogeny, and that such a definition could provide a number of valuable outcomes:

• Firstly, a means to directly compare the predictions of the creationist scenario with those of the evolutionary scenario through an examination of phylogenetic data.

• Secondly, a genuinely testable definition of the creationist term 'kinds' that would allow the unambiguous categorisation of a given organism into a specific kind.

So what would the two scenarios predict? The pictures below, from this site, provide a useful pictorial comparison of the two predictions. 

Figure 1. The predicted evolutionary tree.

Figure 2. The predicted creationist tree.

It is clear from these pictures that phylogenetic analyses examining the distant ancestry of extant species will reveal quantitatively different results depending on which scenario is correct. Under the evolutionary scenario there will be a consistent pattern of similarity forming a nested hierarchical structure stretching as far back as the phylogeny can be resolved (and ultimately as far back as the last universal common ancestor). Under the creationist scenario, on the other hand, the pattern of similarity would only be expected to stretch back as far as the point of the independent creation of kinds. There is no reason to expect there to be any consistent phylogenetic resolution beyond this point.

Thus we have a specific creationist prediction that can be tested - that molecular comparison studies will only obtain a consistent phylogenetic tree as far back as the point of creation, beyond which there will be no further resolvable pattern. As this directly contradicts the evolutionary prediction this provides a definitive means of determining which of the two scenarios is correct

This also provides us with a solid, scientific definition of 'kinds'. If the above scenario was shown to be correct then a kind could be defined as "all of the organisms contained within a consistently resolvable phylogenetic clade". This would correspond to all of the organisms that have been derived from an originally created kind.

I would welcome feedback from creationists and evolutionists on the predictions I have described above. Do they correspond to genuine creationist predictions? If they are falsified, will the creationist paradigm of variation only 'within kind' also be falsified? Anyone wishing to answer these questions can contact me by email.

[ Discussion Forums ] [ Useful Links ] [ Evolutionary Science ] [ Home ]

[ Contact Me ]