ParaphyleticGroup is not a type of MonophyleticGroup

[hlapp]

Otherwise, any paraphyletic clade would by implication also be a monophyletic clade, which is false. (See https://en.wikipedia.org/wiki/Paraphyly, for example.)

Yet, the current version of CDAO asserts this. @arlin do you recall why that choice was made?

[arlin]

The choice was made because this reflects the correct technical meaning of the terms, at least, when I learned them. Monophyletic means having a single origin, i.e., the distinctive features of the group (synapomorphies) only arose once in evolution, and not multiple times. If a monophyletic group includes all of the descendants of the MRCA, it is called "holophyletic" and otherwise it is "paraphyletic."

The locus classicus for these definitions is Ashlock, Systematic Zoology, Vol. 20, No. 1 (Mar., 1971), pp. 63-69.

However, the constant pressure from cladists (who want to win by definition), and naive usage, is to break down this hierarchy and make monophyletic mean "holophyletic." I just googled paraphyletic and looked at about 15 different web sites, and every single one of them presents this incorrectly. And none of them provide any references.

[hlapp]

Thanks @arlin. @ncellinese, do you have thoughts here?

[hlapp]

For Phyloref, I'd like to use the cdao:Clade class in defining kinds of Phyloreferences. cdao:Clade is asserted as a subclass of cdao:Tree and as equivalent to cdao:HolophyleticGroup, which itself is a subclass of cdao:MonophyleticGroup. Since we are not dealing with paraphyletic clades in Phyloref, for the purposes of Phyloref I suppose we could choose to ignore this issue. But we did say we were going to help maintain CDAO.

[hlapp]

BTW @balhoff I suspect you have some opinions here too, don't hesitate to share.

[hlapp]

FYI, added DOI link for the paper that Arlin cites.

[balhoff]

What Arlin is saying makes sense to me. Maybe it’s Wikipedia that should be edited. 😀

[ncellinese]

Sorry, I just saw this thread in my inbox but I feel I am not getting all of the discussion here, so I can’t quite understand what the issue is. Not sure how to comment apart from saying that ParaphyleticGroup should definitively not be a type of MonophyleticGroup regardless on how the concept is applied and to what it applies. I don’t think I am in the cdao mail list.

Nico

On Feb 12, 2019, at 11:23 PM, Jim Balhoff notifications@github.com wrote:

What Arlin is saying makes sense to me. Maybe it’s Wikipedia that should be edited. 😀

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/evoinfo/cdao/issues/15#issuecomment-463052271, or mute the thread https://github.com/notifications/unsubscribe-auth/ACaXwUvFqfuQJvC4yzZB14TJNBUQ1HJfks5vM5M1gaJpZM4awp6U.

[ncellinese]

I clearly didn't follow this conversation. I agree, holophyly is a subclass of monophyly and paraphyly should not be a type of monophyly. Nothing to see here.... :-)

[balhoff]

But I think what Arlin is saying is that if paraphyly is not a subtype of monophyly, there is no point to having a concept 'holophyly'. Otherwise what kinds of monophyly are not holophyly?

[ncellinese]

maybe the correct way would be to keep monophyly and holophyly distinct. These are distinct entities after all, who cares about what the cladists did. However, for phyloref this distinction doesn't matter.

[ncellinese]

Was Arlin trying to suggest that holophyly and paraphyly are the same? If this is the case I disagree. In holophyly, although the ancestor is not included in the group, the group still includes all of the descendants of the same ancestor, which is not the same as a paraphyletic group. So, I confess I am confused by this discussion and can't quite understand who is saying what. Holophyly is a different (Hennigian) way of diagnosing natural entities. Paraphyly do not represent natural entities at all. How off am I now in understanding what you guys are talking about?

[hlapp]

Let's clarify that Arlin's argument is sourced from Ashlock (1971). Ashlock states the following definitions in this paper (emphases as in the quoted original):

1. A monophyletic group is one whose most recent common ancestor is cladistically a member of that group. a. A holophyletic group is a monophyletic group that contains all of the descendants of the most recent common ancestor of that group (monophyly of Hennig). b. A paraphyletic group is a monophyletic group that not contain all of the descendants of the most recent common ancestor of that group.
2. A polyphyletic group is one whose most recent common ancestor is not cladistically a member of that group.

The class hierarchy in CDAO follows Ashlock's definitions. I.e., in CDAO HolophyleticGroup is-a MonophyleticGroup, and ParaphyleticGroup is-a MonophyleticGroup. (is-a = rdfs:subClassOf). It is specifically the latter that is the primary subject of this discussion. Although as @balhoff points out, @arlin adds to that the argument that if we decided to remove that axiom and instead make ParaphyleticGroup a sibling term to MonophyleticGroup, then there is no type of MonophyleticGroup left that isn't a HolophyleticGroup, and then we might as well assert these as equivalent. If we are not comfortable asserting that equivalency, we would in essence be saying there are things that are a MonophyleticGroup but aren't a HolophyleticGroup. So what are these things if we say they can't be a ParaphyleticGroup because a ParaphyleticGroup isn't a type of MonophyleticGroup.

[ncellinese]

I understand what Ashlock said in 1971. In my opinion, we should make ParaphyleticGroup a sibling term to MonophyleticGroup. Also, In my opinion, HolophyleticGroup and MonophyleticGroup are actually different entities. However, If you choose to eliminate holophyletc altogether, and just have ParaphyleticGroup and MonophyleticGroup, I am ok with it, I will live. Simply, my preferred view of Monophyly is very Hennigian and synchronic (all and only descendants of a common ancestor) which is different from the more usual way of defining it (diachronic) as an ancestor and all of its descendants. Again, I will live with what you propose.

[hlapp]

If you think HolophyleticGroup and MonophyleticGroup are not the same, and if we agree that every thing that's a HolophyleticGroup is also a MonophyleticGroup, then it follows that there must be things that are a MonophyleticGroup but are not a HolophyleticGroup. How would you call these things? I.e., of what type are they?

[ncellinese]

The difference is whether the ancestor is considered as part of the group or not. The problem is, as Arlin pointed out, these terms have been applied inconsistently since Hennig. Ashlock tried to clarify Hennig's view but in the process was trumped by the cladists and evolutionary taxonomists and the rest is history. We can choose to ignore all of this and simplify. Just because I (and a few others out there) make a distinction between what I consider monophyly and holophyly doesn't mean we/CDAO needs too. As I said, as long as ParaphyleticGroup is a sibling term to MonophyleticGroup, I am fine with this.

[hlapp]

Thanks @ncellinese. One of the long-standing issues with CDAO is that most terms lack a definition, and a citation. (For example, that the concepts under discussion here and their relationships are based on Ashlock (1971) is entirely absent from the ontology.) So one change I was going to propose at a minimum was to introduce formal definitions and citation(s) for these terms.

I think this should mean that if we rearrange the hierarchy (and possibly make HolophyleticGroup equivalent to MonophyleticGroup), then we should still also propose a formal definition, and if at all possible a citation. Can you suggest a citation for the rearrangement you suggest (i.e., making ParaphyleticGroup a sibling rather than a subclass of MonophyleticGroup), and can you either suggest formal definitions for both of these, or suggest a citation from which to source it.

[ncellinese]

Yes, but I can't really do this right now. I am swamped with mid term exams and a bunch of other stuff and deadlines and I am going to Europe next week. I will make sure to insert this into my to-do list. I'll get it done.

[ncellinese]

you also owe me a paper draft :-)

[hlapp]

I should perhaps also add that with the most recent revision of the Phyloref ontology, a Phyloreference is logically defined as a cdao:Node that cdao:belongs_to a cdao:Clade. CDAO defines cdao:Clade as equivalent to cdao:HolophyleticGroup.

As defined currently, cdao:HolophyleticGroup and cdao:ParaphyleticGroup are siblings. If we moved cdao:ParaphyleticGroup to be a sibling rather than subclass of cdao:MonophyleticGroup, and made cdao:HolophyleticGroup (and thus cdao:Clade) equivalent to cdao:MonophyleticGroup, then the relationship between cdao:ParaphyleticGroup and the concept that a Phyloreference cdao:belongs_to remains the same – namely they have a shared parent concept, and are sibling concepts.

[ncellinese]

I support this scenario

[arlin]

I personally have not conceded to the cladists and their incessant greed to appropriate the terms of debate for themselves. Also, I am not convinced with the argument that "the rest is history." Yes, everyone on the internet thinks that monophyly means holophyly. Everyone on the internet also thinks that chocolate is an aphrodisiac and eating fat makes you fat (to mention 2 food myths). What is popular with the great mass of people is not the same as (1) what is the position of scholarly authorities, and this is again different from (2) what is the position to be taken in an ontology to support reasoning.

[ncellinese]

Monophyly does not mean holophyly, so I am with you here and trying to distinguish between the two. I think we are on the same page, right?

[arlin]

sorry I must have misunderstood. I will re-read more carefully.

[ncellinese]

It's totally understandable because this is fucking confusing for the best of us. The way I define monophyly is sensu Hennig, but he called it holophyly. Ashlock tried to clarify Hennig's view, but the terms were wrongly appropriated, redefined, and the rest is the clusterfuck we know. Am I allowed to swear on GitHub?

[hlapp]

We don't currently have a formal code of conduct for CDAO 😀 I think we can all agree that swearing (or in general disrespectful and unprofessional conduct) directed towards a person or group of persons is not an acceptable form of discourse. By that definition, your swearing so far is fine 😁

[ncellinese]

Phew