GCG 2017 - Classification paper

From systematics

Jump to: navigation, search

Discussion: how we name informal clades, and our first paper

We agree as a group that informal clade names are a help to teaching and to research, to point to a paper that identifies and names the clade if possible. We also agree that we want to get a paper out relatively soon that moves the process forward, gives us some working names for clades. In this paper, we allow for a mixture of formal and informal names. No big long introductions, just laying out the framework for ongoing work. With the right kind of introduction, this paper can also serve as a model for other genera. We also propose that the clades have some morphological characters associated with the informal clades -- in the framework paper, for the informal and formal groups, we provide a little discussion about distribution and morphological tendencies.... or we confess that it doesn't have morphological coherence if we can't find any. Additionally, we should explain the scale at which we are parsing the tree: we are using clades that are identifiable, reasonable in size, hopefully morphologically identifiable, but a unit that we can bite off.

Here is a risk of doing this in a paper: it's static. We need a dynamic platform for updating this, and the Scratchpad may be a very appropriate place.

Question: will the paper have a tree or a list or both? If a tree, what tree?

Tree 
In the paper there should be a tree; then there should be an appendix with a list of all the spp that we think is part of each group. For the tree, as we're putting the paper together, we could take the 2016 phylogeny as the tree we are using, or we could put together a representative stylized paper that lays out the trees as far as we know, just lay out what the guide framework is. We don't try to present one ML tree and make that be the foundation of what we do, because no single analysis represents our knowledge on the genus. The big tree we are working on is just a part of our evidence: we don't intend to present that paper as a whole.
Species list 
The list should include the entire list of species as best we know it, and we place species as best we can. We explicitly mark those that have been published in a phylogeny, and we are not afraid to acknowledge that we don't know for sure where a species goes. It will be very important to be clear about what we know and what we don't know. It's immensely important that we lay out what we don't know, because anything we write may be taken as the final truth. It's useful to lay out that we don't know everything.

Should we be using common names or Latin names for informal clades? Argument for Latin: the common names sound bizarre in other languages. Argument for common names: the use of Latin names will make it sound as though we are making nomenclatural recommendations before we are, prematurely. We could also use "alliance" with the Latin names, so that we have some consistency. We should have some consistency. Or we could use "clade" or "lineage." Informal has also the sense of "familiar."

We need to outline these principles in the framework classification paper. We need to be in agreement that what we propose is the right path forward. This discussion should aim at a GCG paper that (1) lays out principles of classification and (2) outlines a number of informal names.

What if we pull the section name out and just use the species name, followed by "Clade"? The problem will be selecting a species... possibly a species that sections are based on because those names are already well known. Where possible, using one of the oldest names (or a type name)... if we have information on priority, we can default to using that name. Because they are informal names, we don't have to be married to priority. But also consider that we want it to be usable, so it might make sense to use a common widespread species epithet. There was discussion on how important priority is, whether we want a set of rules or whether we want a set of guiding principles. The benefit of agreeing on these principles now will help when we are working remotely.

Concern about using clade / lineage: if we do this, then we may have to recant when we find that this or that is not a clade. On the flip side, everything we are making claims about in this group we believe to be a clade. We may be wrong about some in the future, and we may need to recant no matter what we do. Also, from a teaching perspective, does clade cause problems because you can define a clade in many ways? We discussed PhyloCode, but there seems to be consensus that we don't want to be hamstrung by clade definitions when in fact the tree is still in flux.

Another option: use a shortened version of the section name (e.g., "Vesi clade") with a common name.

Our approach: Taking published phylogenies (GCG 2016 + everything else... anything that clarifies lineages), using those to identify informal clade names and incertae sedis spp, and formal clade names where we can easily do so, and presenting a GCG framework paper. The principles by which we come up with names: we start with familiar species, often types of sections within the clade, with the caveat that we want names that are user-friendly and not readily confused with other species.

Working on the paper: Eric and Andrew will take a first crack at it. Santi has already put together a first framework tree with triangles for clades, which we could edit and circulate. We also need to commit people to work on certain groups. We could start today ... we need to bring people in from the outside on various groups, people who are not in the meeting today but whose expertise we need. Question: how do you go from these informal categories to a formal classification.

Timing and mechanics of first paper

We propose getting a draft (Eric and Andrew) to the group before a late February conference call, then conference calls in late March and April, leading to a late May submission, perhaps to Bot J Linn Soc. Goals of the paper:

  1. Recognize subgenera -- probably the only clade that needs a names is Siderosticta / Hypolytroides. The subgenera would be
    1. core Carex
    2. siderosticta lineage
    3. Vignea
    4. caricoid, or unispicate + Schoenoxiphium
  2. Recognize informal clade names
  3. Clean up what sections we can

One concern: if we define subgenera, do we paint ourselves into a corner? A: We get out of this by letting the informal names intervene between the subgenus and sectional levels.

Coordinated papers:

  1. Big phylogeny (Santi)
  2. Follow-up paper on Vignea (Roalson, Hipp, Waterway)

Schedule of meetings and homework assignments

Personal tools