User blog:Cerne/New taxonomic system

In my final oversized section for the Part 2 update, I will describe the taxonomic classification system that I plan to use for the organisms living in my conworld. It is a personal method; I do not expect anyone else to use it or want to use it, and I am not suggesting that it is in any way superior to the classification systems currently being used in evolutionary and taxonomic science. In addition, I am not claiming credit for originality; I do not know whether the method I am using has been thought of wholely or in part by other people before me or not, and if it has, then I am not aware of it.

This classification system actually came from something I typed a number of years ago that was a lot more extensive and specific but it had never been posted anywhere. It had phylogenies, descriptions of evolutionary patterns and forms of speciation, taxon profiles with primitive and derived traits (called "characters" in taxonomy), and more. So, yeah, there is more to it than what is in this entry, but I am only putting forth an outline and providing a basic description of how the whole thing works. There are pages of notes somewhere that I wrote for it so if I do type any Wikia articles on this classification system then I will try to include them as much as I can.

In the meantime, here's what I've got for an introduction.

......

When inventing new kinds of organisms for my conworld, I obviously need a way to classify them and represent each species' phylogenetic relationships with one another. At one time I did what I saw just about every other conworlder and speculative biologist did: I used Greek or Latin binomials and Linnaean taxonomic nomenclature. Or at least I tried to. At first it worked well, but then I came across some issues that made me question the whole practice altogether.

This post from the Spec-Evo site sums it up pretty well:

=
I no longer bother with Greek and Latin nomenclature. For that matter, I don't even bother with Linnean classification. My evolutionary taxonomies are purely phylogenetic. I.e. One species descended from another species, descended from another species, and so on and so forth. I've even redefined what a "species" is on the fictional planet I am working on right now. Basically, I'll have Regnum which is the group of organisms I am displaying the phylogeny for, like "vertebrates" or "tetrapods" for instance, and that will be followed by succession of Phylum all the way down to the species level. Then, I work my way up from Species, Genus, then Family. No Class or Order. So Regnum is a LOT more relativistic. Family and Genus will be more phenetic than genetic, and they - along with the names of individual species - will be in the fictional language(s) I've invented. Molecular "clocks" in the DNA can be used, along with the new definition of Species, to construct the secession of species only.

I am actually thinking of designating different species and secessions with some sort of number code that follows the order of divergence from the basal species, followed by the common species name, then the crown group, then a short description. Quite frankly, the whole thing seems kind of overwhelming to me right now. My planet needs work before I can re-classify everything living on it.

So why did I choose this road? Why couldn't I settle with Latin binomials and Linnean nomenclature?

To be honest, the idea of thinking up Latin and Greek binomials became more problematic than what I am doing right now. I had a hard time finding good binomials that weren't already chosen, all the GOOD words were already used, and trying to find ANY classical Greek words to use was just getting too difficult. So I gave up. Plus, if you think about it, all of these Greek and Latin binomials are EARTH names, after all. Who says a sophont species on another world has to go about biological classification in the same way? They'll be using their own words anyway, won't they?

So, yeah, none of those fancy Greek and Latin names for me=====

--SNIP--
I may also have described my issues with using Linnaean nomenclature in a previous entry here, but if I did then I don't remember it very well. Namely, it got too restrictive. If you think about it, a seven-taxon system is not going to cover every single taxon we've come up with thus far for any group of organisms on Earth. Even after adding all these extra taxa, like "Sub-Class," "Infra-Order," "Super-Family," "Tribe," etc., I still find many taxa "unranked" and others simply unaccounted for, without even a temporary filler name for that spot in the taxonomy.

The first thing I thought of doing is getting rid of the Class and Order ranks, as well as any additional ranks that use the suffixes "sub-," "supra-/super" and "infra-," and making the Kingdom taxon (now strictly called "Regnum") relative to the group of species being featured. "Species" still means what it has always meant, but instead of determining it on which populations can and cannot breed with one another, I choose to determine it on which species do and do not breed with each other under normal circumstances. At least then I can avoid the ongoing dispute on what a species actually is whenever a new hybrid animal is discovered.

Anyway, from that, I match each species with its closest relative to come up with groups; Genus being first up and representing populations that can no longer have viable offspring with those outside of the group, then Family representing groups that are very similar genetically as well as phenetically. The latter is for the most part superficial and some genera may be the sole member of their taxonomic family if they are physiologically distinct enough. From there, each group is called a Phylum and each phylum can contain - and be contained by - other phyla. Every group of species is called a "phylum," actually, but "genus" is kept for a practical purpose and "Family" is a remnant of my earlier bought with Linnaean taxonomy that I kept for aesthetic reasons.

So, upon establishing what species and phyla are, one must then think about how to organize them. It should now be expected that each phylum is represented by a single species and contains all of its descendents. What I do is represent each successive phylum as a species and then follow it with a group name. For example: for a while Tiktaalik was considered to be the ancestor of all terrestrial tetrapods (I don't know if it still is or not) so its phylum representation might go something like "Tiktaalik / Tetrapoda." Each phylum's species representation can be thought of as its basal species when counted amongst its descendents so this is what I call the first part of each phylum representation. When listing descendants, the basal species can go first or last, depending on whether I decide to list descendant species by order of divergence or genetic closeness to the parent species. After that, each subsequent species can be listed by order of divergence.

The type of adaptation must also be kept in mind. When devising this current scheme, I had speculated that there was a certain order in the types of adaptations that would be made, according to what the organism was adapting to. Favorable conditions would be followed by an adaptive radiation while unfavorable conditions would be followed by more extreme adaptations that would sometimes lead to exaptation. An existing trait or "character" would be further selected for and enhanced in favourable conditions, and new traits would be selected for during unfavorable conditions, like a sudden change in environment. If this can be indicated in any way in the designation for each species, it would then be possible to see what kinds of changes happened in the parent species' environment that led to the existence of each daughter species.

Further attention can be paid to what are called "novel adaptations" and what I like to call "innovations." Specifically, the novel adaptations are those adaptations that either utilize an existing trait for what it is adapted to do or utilize a new trait for something the organism wasn't doing before. "Innovations" are basically instances of convergent evolution where a "sister" species attempts to do what has already been done but is unable to with the same trait and thus needs to make the same adaptation with a different trait. I believe there is a terminological(?) distinction between the two; whereas anatomical convergence involves the same adaptation appearing in the same part of the body (I.e. the wings of bats and pterosaurs), functional convergence involves the same adaptations appearing in different parts of the body (I.e. vertebrate wings and insect wings).

The idea behind this observation comes from Dollo's Law Of Irreversibility where - once two species diverge - the species that didn't make the adaptation in question is never able to make the exact same adaptation again and therefore needs to make a very similar adaptation in a different way. So there is a trade-off, but also an exception: if the adaptation in question was not the cause of the divergence (I.e. it evolved after the divergence occurred) then it can happen again in the same way.

And then there is the phenomenon of genetic back-tracking, where a trait that has been lost in an organism suddenly re-appears. This is mostly done through artificial selection whereby a hybrid organism is bred back into one of its parents' lineages and the resulting offspring acquires older ancestral traits through the deactivation of certain "genetic switches" (see: NOVA - What Darwin Never Knew), but apparently this sort of thing also occurs in the wild.

The paradoxically funny thing is that theoretically, if this holds true, the more distantly related species may more likely resemble each other than those that are more closely related. But this would only be true if both species have the same body part to adapt for the same reason. And remember that it is only a hypothetical supposition. I shouldn't need to say I may be wrong but I may very well be. Just so you know.

ANYWAY...

What all of this means is that if unfavorable conditions do occur, like rapid environmental change, for instance, the extent to which it happens can be shown in the phylogeny. Of course that all depends on how important the trait in question actually is in adapting to those conditions, so in determining that fact it is important to note just how often the adaptation shows up within a phylum and across closely related phyla. I plan to indicate this somehow, if not in the species/phylum label itself then maybe in a profile of the species or phylum.

Now, words may be good enough to a certain extent but I should also show how this is all going to work out on a cladogram or some other type of phylogeny diagram. The approach I will be taking is adverse to what is traditionally done in textbooks about evolution and taxonomy (see facing image).

Taxonomy as it is represented in the literature is very hierarchical; the taxa that are more "advanced" or that have gone through more changes are represented as a straight lineage whereas those taxa that did not change are mere offshoots. It should be no surprise, then, to find that taxonomic diagrams of human evolution nearly always tend to have our own lineage as straight as possible. The changes in traits for the species are indicated by a small dash to indicate when that particular adaptation took place.

When I say "adverse," I am talking about representing the basal species in a phylum as a straight lineage and representing divergences as offshoots. The result won't be a clean, clear-cut triangle or bell-shaped diagram. Rather, it may look more like a straggly bush. But I feel this may be a more accurate portrayal of taxonomic relationships in a given phylum or collection of succeeding phyla. It is more equal, emphasizes the adaptations that are less noticed, and treats each and every organism as a novel organism in its own right, rather than a "lesser" form of something else.

Here is a demonstration of a phylum with a lot of divergences:



And here is an example of a collection of subsequent phyla with an extremely minimalistic number of divergences per lineage:



An average diagram would be a mix of the two but would probably take up considerably more room so I won't display one here. Instead, I will use something like this: Species 1 |   +--> Species 2 |      |    |       `--> Species 4 |   `--> Species 3 NOTE: Basal species are not represented here because it is assumed they are being represented by the parent species. This is done due to the limitations of strictly using text to display phylogeny. However, it would be possible to include the basal species as an extension of the first vertically represented lineage.

In addition to the species/phylum representation, each species and phylum label will include a numerical code to indicate in what order they diverged from their parent species. To give you an example using the above diagram: Species 1 --> Species 1.1 (parent species) \-> Species 1.2 (basal daughter species) Species 2 --> Species 2.1 Species 3 --> Species 3.2 Species 4 --> Species 4.1 The larger the diagram, the more numerals you would need. The amount of divergences, as indicated by the second number, and the order in which the divergences took place, as indicated by the first number, should together be able to provide an accurate enough depiction of whether there was an adaptive radiation or a lot of natural selection taking place at the time of divergence. I am thinking about finding a way to add an indication of trait novelty or innovation somehow but I'm not going to get into that right now. If I go any further, this entry may never get published.

One thing very worth noting, and something I went into with reasonable detail in that quote I brought up at the beginning of the entry, is the actual name of the species. Greek and Latin binomials don't work for me. They never did, though I did use them for a while. The problem I had with them was simple: they were already being used. More specifically, they were a system that used Earth languages to name species on Earth. Hundreds if not thousands of useful and relevant binomials are already being used. Nevermind the difficulty of coming up with new binomials; the chances of coming up with a good binomial that has not yet been used is surprisingly slim. And they are becoming slimmer. I cannot begin to tell you how many times I have come up with a decently satisfactory binomial only to find out that part or all of it has already been used, and that I need to change it without changing the meaning of the binomial or find a different alternative. And, really, it makes no sense; why am I going out of my way to find a name for an organism on another planet in a language that is only found on Earth when - chances are - it won't even be used on that planet anyway? So not only are Greek and Latin binomials too restrictive, they are also impractical and unnecessary. Why, then, would you need to use a method of naming species that is only relevant to you when you can use a method that is relevant to everyone?

When I was younger and just beginning to dabble in fictional and speculative biology, I would give my creatures these weird names that were either completely fictional - maybe in some fictional language that I never bothered to develop beyond that name - or were a portmanteau of English words used to describe the species. A good example is "Gremog" which is a squat, roundish, dog-like creature with the ears and short face of a gremlin. But I always liked the simpler method of naming creatures for what they did because those names were so versatile. Take "Side-hill Gouger," for instance: you can translate that into just about any language on Earth, even Greek and Latin, and you don't even need to look it up in a foreign language dictionary. You can translate the name just as easily into a fictional alien language. BTW, "side-hill gouger" was a name my grandfather came up with for gophers or some other Earth mammal that dug holes on the sides of hills.

So what I do here, after applying the numeral code to the parent-species part of the phylum label, is follow it with a short description or portmanteau in English that describes it. Eventually I plant to have a similar naming scheme in Shixa which is the naming language I will be using to name and classify other things like place names and individuals' names. What I will probably do is use a "folk name" for the species part of the label and a language-relative description or portmanteau for the phylum part of the label.

As for the phylum part, here is how I think about it: a species diverges from its parent species because of some adaptation it has acquired but that none of its sister species have acquired, so it becomes a matter of "this" and "everything else." In taxonomy, this is called monophyly and paraphyly. What I plan to do is use monophyly and paraphyly in a more relative context; every sister species in a phylum has something in common, even though each of them has made their own adaptations and each is therefore unique, so they can still be classified paraphyletically as well as monophyletically. What the phylum representation in the species label does is classify each of its descendents paraphyletically. What the species representation does is classify that species monophyletically. In conjunction, these two representations can allow you to define and label a particular species without resorting to a rigid "in-group/out-group" dynamic that - in taxonomy right now - only seems to apply to one species only. That's why taxonomists don't like paraphyly very much: it misidentifies species that don't have the adapted trait they want to focus on. I am saying that paraphyly can correctly identify a species as being part of a genetically-related group as long as it is used in a relative sense and in conjunction with monophyly, the latter of which should actually be applied to all sister taxa. That is, we should be trying to find out what makes each and every taxon unique. If applied correctly, this way of representing species may even get rid of the divide between "lumpers" and "splitters."

......

That will be it for now. And this is all I am going to type in no less than a week. To be honest, the whole thing was a bit exhaustive even though I had an easier time knowing what I wanted to say. I want to get more updates in but these will be actual updates and not a series of small sections that grow larger and larger. Yet another learning curve, I guess.

Thanks for reading. Imma go to bed now.