Thoughts: How many concepts are we talking about
Fourth post in this sequence (here are posts 1, 2, 3, respectively). Changing gears a little. The motivation for this post is to explore the interactions of explicitly and implicitly communicated taxonomic concepts in conversations among (living, meeting) humans with comparable levels of taxonomic expertise. How many identifiers are we talking about?
The exploration has two parts. The first part simulates a brief conversation of the kind that two human speakers may engage in while meeting in the hallways at a taxonomically oriented conference. The speakers know of each other, either through prior personal interactions or (minimally) by having read several of each other’s taxonomic publications. The conversation is hypothetical, and even though certain real persons are mentioned, the sole purpose of this is to add some realism, not to pass my judgment on any taxonomic particulars. The post is about exploring how the issue of taxonomic name/concept identifier resolution relates to this kind of communication, generally.
The second part examines the conservation from the perspective of representing taxonomic reference – “logically”. By that I mean framing the taxonomic content identifiers communicated explicitly or implicitly by the human speakers in such a way that a computational, logic-based application can adequately represent them. Ok, so here goes (in part, as it will turn out).
Two beetle taxonomists – Betty and Sofia – meet in the hallways of the Annual Meeting of the Entomological Society of America.
Betty: Oh hi, Sofia, great to see you at Ent. Soc. How have you been?
Sofia: Hi Betty, thank you, quite well actually. Say, did you happen to see Natalie’s talk this morning on her new baridine weevil classification? She just published on this in ZooKeys.
Betty: Yep, I was there, back left corner. Seen the paper too, came out last month. Quite the number of changes there.
Sofia: I’ll say. Did you notice what she did with the baridine tribes? Madarines and madopterines are all mixed up now.
Betty: Well, of course. They’ve been a mess ever since Casey’s Memoirs got published.
Sofia: Right, right. Mind you, the Catalogue couldn’t resolve everything either.
Betty: You mean O’Brien & Wibmer?
Sofia: No, the 1999 Catalogue. Remember, it got published just prior to the new Code coming out and made lots of changes, or at least clarifications.
Betty:. Got you. We keep plugging away. What do you think of the recent molecular work?
Sofia: Regarding baridine tribal relationships?
Betty: Or even more generally.
Sofia: I see. Well, I think that the sampling remains too spotty. And the most recent studies no longer include morphological characters, so we cannot understand how these new presumed curculionine, molytine, conoderine, baradine groupings may be diagnosed.
Betty: I still follow the Catalogue for higher-level, primarily because it is explicit and comprehensive. Except I use ZooKeys – Bouchard et al. as an update above the genus level.
Sofia: Then how do you feel about the “baridine” chapter in the Handbook?
Betty: Interesting! I believe it’s Bariditae now? Inside the conoderines. I like that classification, reminds me of that neat anatomy paper by Korotyaev and co-authors in the Washington Ent. Soc. Proceedings. Anyway, lots of new names to learn..as always. I hope this work gets continued to address baridine.., I mean, bariditine, tribes and subtribes.
Sofia: One can only hope. Got to run, Betty. Would love to catch up more later. Will you be at the mixer?
Betty: Yes. Let’s also chat with Natalie then, see what she thinks about the Handbook classification in relation to hers.
And so concludes the hallway discussion between Betty and Sofia. On to the analysis stage. Notice that I colored several mentions of (colloquial approximations of) taxonomic names in maroon, and mentions of (colloquial approximations of) taxonomic sources in indigo. As an initial exercise to make sense of this conversation, we can list the taxonomic same strings, as is and in alphabetical order.
- baradine (groupings, subtribes, tribal relationships, tribes, weevil), “baradine”, Bariditae, bariditine (subtribes, tribes), conoderine(s), curculionine, madarines, madopterines, molytine.
I think a possible human expert reading of the above list of name strings might assess that at least the following valid Latin taxonomic names – as referenced in Alonso-Zarazaga & Lyal (1999) and Prena et al. (2014) [Bariditae] – are in play:
- Baridinae Schoenherr, 1836 [subfamily]
- Bariditae Schoenherr, 1836 [supertribe]
- Conoderinae Schoenherr, 1833 [subfamily]
- Curculioninae Latreille, 1802 [subfamily]
- Madarini Jekel, 1865 [tribe]
- Madopterini Lacordaire, 1866 [tribe]
- Molytinae Schoenherr, 1823 [subfamily]
So this would translate into minimally eight taxonomic names. How many sources might be referenced in connection to these taxonomic names? Again, we apply the inference powers of a thoroughly informed human taxonomic expert.
- Catalogue. That would be (not including the 2002 and 2006 “Addenda and corrigenda”): Alonso-Zarazaga, M.A. & C.H.C. Lyal. 1999. A World Catalogue of Families and Genera of Curculionoidea (Insecta: Coleoptera) (Excepting Scolytidae and Platypodidae). Entomopraxis, Barcelona. Source abbreviation: sec. Alonso-Zarazaga & Lyal (1999).
- Bouchard et al. That would be: Bouchard, P., Y. Bousquet, A.E. Davies, M.A. Alonso-Zarazaga, J.F. Lawrence, C.H.C. Lyal, A.F. Newton, C.A.M. Reid, M. Schmitt, S.A. Ślipiński & A.B.T. Smith. 2011. Family-group names in Coleoptera (Insecta). ZooKeys 88: 1–972. Source abbreviation: sec. Bouchard et al. (2011).
- Casey’s Memoirs. This most likely refers to (not accounting for Casey 1892, 1893, 1920 – the latter is also part of the Memoirs but largely avoids reference to higher-level taxonomic entities): Casey, T.L. 1922. Studies in the Rhyncophorous subfamily Barinae of the Brazilian fauna. Memoirs on the Coleoptera 10: 1–520. Source abbreviation: sec. Casey (1922).
- Natalie in ZooKeys. This is a hypothetical publication (*), assuming that Natalie’s presentation strictly adheres to that publication. Let’s reference it as González, N. 2015. A revised, phylogenetic classification of the baridine weevil tribes (Coleoptera: Curculionidae). ZooKeys 500*: 1-50. Source abbreviation: sec. González (2015).
- Korotyaev and co-authors. This is, in context: Korotyaev, B.A., A.S. Konstantinov & C.W. O’Brien. 2000. A new genus of the Orobitidinae and discussion of its relationships (Coleoptera: Curculionidae). Proceedings of the Entomological Society of Washington 102: 929-956. Source abbreviation: sec. Korotyaev et al. (2000).
- Recent molecular work. This is to my mind the least explicit but nevertheless partially resolvable reference. Several potentially intended sources come to mind, of which I will choose: McKenna, D.D., A.S. Sequeira, A.E. Marvaldi & B.D. Farrell. Temporal lags and overlap in the diversification of weevils and flowering plants. Proceedings of the National Academy of Sciences of the United States of America 106: 7083–7088. This is not the only plausible choice, but it fits the two speakers’ contextual intentions of referencing a strictly molecular analysis in which certain traditionally recognized weevil subfamily-level taxonomic concepts are not monophyletically resolved. Source abbreviation: sec. McKenna et al. (2009).
- O’Brien & Wibmer. That should be, mostly immediately (not accounting for the South American complement Wibmer & O’Brien 1986; or Supplements [1984, 1989]): O’Brien, C.W. & G.J. Wibmer. 1982. Annotated checklist of the weevils (Curculionidae sensu lato) of North America, Central America, and the West Indies (Coleoptera: Curculionoidea). Memoirs of the American Entomological Institute 34: 1–382. Source abbreviation: sec. O’Brien & Wibmer (1982).
- Handbook. That is: Prena, J., E. Colonnelli & H.A. Hespenheide. 2014. Conoderinae Schoenherr, 1833; pp. 577-589. In: Leschen, R.A.B. & R.G. Beutel, Editors. Handbook of Zoology. Arthropoda: Insecta: Coleoptera, Beetles. Volume 3: Morphology and Systematics (Phytophaga). de Gruyter, Berlin. Source abbreviation: sec. Prena et al. 2014.
So then we have (again, minimally) eight taxonomic sources that Betty and Sofia are jointly referring to. Eight names times eight sources would translate maximally into 64 taxonomic concept labels of the syntax: taxonomic name sec. taxonomic source.
My point here is about identifiers for communicating individually and jointly perceived taxonomic content. Betty and Sofia’s conversation can be read as an iterative taxonomic identifier/meaning communication and resolution game. On the surface, perhaps only eight identifiers – the explicitly stated taxonomic names – are in play. However, just by virtue of Betty and Sofia coming together (while both are alive and mentally competent and trained in comparable taxonomic legacies) and recognizing each other’s expertise, all taxonomic names are at least implicitly contextualized. Their conversation does not occur in any conceivable taxonomic semantics framework to equal measure. For instance, if Betty faced Sofia in the 2015 meeting hallways and uttered nothing except “Baridinae”, there is a higher likelihood that both Betty and Sofia would internally augment the resolution of this identifier to symbolize (something quite close to) the meaning of the taxonomic concept label Baridinae sec. Alonso-Zarazaga & Lyal (1999) – this being the most recent and globally comprehensive treatment of weevil mid-level taxonomy that both are ‘conditioned on’ – than to symbolize the meaning of (something quite close to) Baridinae (really: Barididae) sec. Schoenherr 1836. This may seem trivial if you (dear blog reader) happen to be a human whose brain shares a lot of characteristics with those of Betty and Sofia. ‘We’ typically don’t walk around uttering taxonomic names with strong intentions to identify taxonomic meanings that we jointly ‘know’ to be outdated. We also don’t usually operate on the assumption that successful contemporary discourse about taxonomic content depends on each speaker promoting meanings via identical name strings that are maximally different from the other speaker, or strive deploy other ‘unintuitive’ assumptions. Instead we want to understand each other, and are conditioned to adhere (albeit with some degree of variation) to a number of inter-subjectively shared, actionable framing assumptions that help realize this motivation.
Acting reliably on these human motivations and assumptions related to successful taxonomic content identification and communication is not trivial for computational logic. Sometimes this challenge is described under the term Frame Problem, roughly: how can holistic, open-ended, context-sensitive relevance (in human-to-human communication) be captured by a set of propositional, language-like representations of the sort used in computational logic? In the ‘eyes’ of a computational representation, the uttering of “Baridinae” during Betty and Sofia’s hallway meeting identifies any and all taxonomic meanings that this taxonomic name has and will symbolize in the open-ended legacy of human taxonomy making. Without additional, explicit contextualization, the taxonomic name symbolizes an entire, open-ended taxonomic concept lineage to the computational representation (note: this is not all that that the name may logically represent). And indeed, we can ascribe to Betty and Sofia that they are providing identifiers for eight such lineages to a logic-enabled environment.
To a significant degree, Betty and Sofia succeed in understanding each other in the above conversation. They do so by drawing on their shared inferential abilities to augment the identifier resolution directly facilitated by the taxonomic names they use to the level of implicitly contextualized taxonomic concept labels. They thereby effectively address and solve the Frame Problem. This also means that the actual number of taxonomic concept identifiers they need coin (internally, inter-subjectively) in order to realize their communication motivations is higher than eight, but presumably lower than 64. Where ‘exactly’ they lie within that identifier range will be explored in a subsequent post. Lastly, we can hold on to the idea that Betty and Sofia’s Frame Problem solution is not effectively communicated to the computational logic realm just by the conventions of their human-to-human expert taxonomist conversation. Look, mom, I almost managed to end this post with the world “realm”.