Archive for March, 2020

Thoughts on TaxonRelationshipAssertion

Issue: More appropriate name for TaxonRelationshipAssertion

(Message composed for the taxon names and concepts interest group; too long to put into the github comment stream, so depositing it here so I can just drop a link.)

TL;DR the vocabulary as it stands is mostly OK, we just need to clarify the documentation.

I didn’t read every word so I apologize if I repeat someone or have missed some important point.

I looked at the draft specification (? the tnu_terms table) and I think it is workable; it just needs to have the underlying theory tightened up a little so that readers are steered away from confusion.

I think the idea of a TaxonomicNameUsage, and that of a TaxonomicName, are excellent.

I think the documentation for parentNameUsage, vernacularName, and preferredName all need to be clarified to emphasize that this information is according to the what the source says (NOT according to the author, who may have changed her/his mind since writing the source!). We need to be very clear that the purpose of this class is to anchor what we say to documentary evidence, and to draw a line between what the source says and how we interpret it. If we want to interpret, we will do so in sources we write, and that will lead to our own TNUs.

The definition of TNU (“operationalization of a taxonomic concept”) is vague and leans too heavily on “taxonomic concept”. It doesn’t help us to know what these things are like, what their properties are, what might be true of them (these are the basic questions of ontology). Morphologically a TNU would have to be a kind of usage. Webster’s says a usage is a generally accepted practice or procedure – an action that somebody takes with something. So one might say “that usage is familiar / unfamiliar / obsolete” (potential properties). Or it “is common in the UK” or “is no longer in vogue”. And we can assume that what is being used is some ‘verbatim name string’. TNU is slightly more granular that this in that TNU 1 and TNU 2, both of the same verbatim name string, might have indistinguishable usage, yet still be different TNUs because the sources are different. This to me does not seem very consequential because we can always imagine that the two references lead to some difference in usage, even if it’s undetectable in the population of ‘users’.

This takes us to TaxonRelationshipAssertion. The word ‘taxon’ has a fraught history and I understand why biodiversity informaticians treat it as radioactive. I read Lam’s paper “What is a taxon” – he claims responsibility for reviving the word in the 1950s, after it languished for twenty-five years or so – and he gives a series of definitions that are wildly in contradiction. He finally settles on saying it has to mean whatever the ICZN code says it means, and that it should not be used in botany.

I am not opposed to using the word ‘taxon’. It is a delightful word if you ignore its baggage. To me the best definition is that taxa are the subject matter of taxonomy, where taxonomy is taken in the sense of biological classification, with nomenclature being an independent pursuit. That is, I want ‘taxon’ to be a biological entity, not a human, administrative entity. We have ‘name’ and ‘TNU’ as good administrative entities, but when you do science you have to interpret names or TNUs as biological entities.

In particular, we don’t want the situation where a group for millions of years is not a taxon, and then suddenly, when an article describing it is published, it becomes a taxon. Or, to have a manuscript submitted for review be rejected for calling a group a taxon, just because there has not yet been a publication that describes the group.

So, nothing that a human can do (other than ecological manipulation such as extirpation) should be able to affect anything we say about a taxon (in this sense).

Because I will probably be attacked or misunderstood for trying to define ‘taxon’ this way (as a biological entity), I will, for now, just use the word ‘group’. (However… I see that the ‘TNU terms’ table seems to use ‘taxon’ in just this way – see definition of TaxonRelationshipAssertion. If people here like ‘taxon’ I celebrate. anyhow.)

By ‘group’ I don’t mean the mathematical notion of ‘set’, or Lam’s other candidate meaning of natural grouping based on characters; I mean a group that a competent taxonomist might circumscribe. I don’t know if we can, or need to be, more precise than that.

Given a word for the biological groupings we care about, this lets me talk about ‘TaxonRelationshipAssertion’.

What is going on is not that TNUs are groups, but that they are interpreted as designating groups. We want to be able to claim that if t1 and t2 are TNUs, then the groups that we interpret t1 and t2 to be are equivalent, or satisfy some other RCC-5 relation, etc. etc. It is not the TNUs that are equivalent or whatever, it is the groups.

We don’t need to put groups into the vocabulary to accomplish this. We just have to change the documentation of some terms to make it clear what’s going on.

And by the way the word ‘concept’ has no place here at all. Concepts are theoretical entities that live inside human minds – or perhaps in linguistic communities, who knows – nobody really knows what they are. Luckily we don’t need know. Our subject matter is not psychology or linguistics, and does not include concepts.

OK, we were talking how to express a claim that two groups (which might actually be only one group) are related in some way. I agree with Markus that ‘assertion’ is redundant: we assert what we believe (conjecture, calculate, repeat, etc.) just by saying it. The important thing here is not the asserting, which is ubiquitous in the document at hand, but the relationship between the groups. If we document that the groups in question (those whose relationship we’re claiming) are the ones that the subjectTNU and objectTNU are *about*, that the TNUs’ purpose is to tell us which groups we’re talking about, then everything will be clearer.

(I have taken to use a ‘relationship’ as something that holds between two particular individuals, and a ‘relation’ to be a general pattern connecting individuals of various classes. This is a technical distinction that is probaby not in widespread use.)

So ‘group’ would only occur in definition strings, and possibly in GroupRelationship (or TaxonRelationship, or whatever term you/we choose as long as it has nothing to do with names or concepts).

Now I started this all off by saying that TaxonRelationshipAssertion was not a good name because it suggested a relationship between taxa, rather than between TNUs. I contradict myself and I am sorry. The relationship is ultimately between taxa/groups, but it is expressed with the assistance of the TNUs, and therefore induces an incidental relationship between the TNUs. Suppose t and u are TNUs, and I is the interpretation function that takes TNUs to groups (that is: I(t) is the group that t is about). Then there really are two relationships, say R1 and R2, where R1 holds between the TNUs R1(t,u), and R2 holds between the groups R2(I(t), I(u)). I was thinking of R1 when I complained, but the purpose of the ‘assertion’ is to make an R2-type claim, so that is why I’m now thinking it’s OK for the connective to talk about the groups/taxa instead of the TNUs.

This doesn’t preclude the addition of Group or Taxon as a class in the vocabulary. I don’t have an opinion yet on whether that would be desirable. I know it’s been discussed in this comment thread but I have not followed the argument in detail.

Addendum about ‘circumscription’: A circumscription is an act of circumscribing, and circumscribing is something that people do to help other people know what group (or similar bounded entity) they are talking about. So circumscriptions are not groups, and do not belong to the realm of biology. If we look at a piece of writing (or a video of a lecture, etc.) we may find words that we would call a ‘circumscription’, but that is not quite accurate – it is just a very common metonymy where we confuse some words with what they say. A similar case is ‘contract’ where most people freely mix up the contract itself, which is an agreement or ‘meeting of minds’, with its written record. These are different things with different properties. Closely related, yes, but not the same.

In writing all this I’m trying to set an example of a certain way of talking about and choosing words. In technical writing, in my opinion, it is best to avoid metonymy and to agree with ordinary language, unless there’s a very good reason not to (and from time to time there is). Avoiding metonymy means to use different words for different things. Consistent with broader usage in society ought to be easy to judge, if not to do. It often helps to just look at a dictionary, but in difficult cases a corpus analysis or literature review might be called for.

Categories: Uncategorized