On Earth, we use taxonomy to organize biological organisms into an identifiable hierarchy in which organisms that are evolutionarily related are placed in close proximity. This evolutionary approach to classification is a fairly recent innovation, dating back to Charles Darwin's publication of "Origin of Species."

As we Earthlings begin to move away from our little life-bearing planet, we will no doubt begin to discover organisms, living or dead, on foreign worlds (the odds are in our favor that life exists/existed somewhere other than Earth). Humans, being human, like order and like things to be neatly sorted. These new organisms would almost certainly go through the taxonomic system to be placed somewhere in the Hierarchy of Life (copyright pending), but here we face a complication.

Extraterrestrial life is highly unlikely to have evolved in the same manner as Earth life, and with even less likelihood of having a common ancestor. Our current taxonomic system utilizes evolution as a key factor in classification, but we can't know the evolution of extraterrestrial life without spending years doing field and lab work to determine how organisms are related, and short-lived, fast-paced humans want answers now.

Enter xenotaxonomy: the science of categorizing extraterrestrial life. On a given world, evolution may have evolved creatures unlike anything on Earth, or maybe creatures like we see in our myths, but not in real life: dragons, giant sea serpents, griffons, etc.

What would an effective xenotaxanomic system look like, given the desire both to have answers quickly and for the resulting hierarchy to be clear and stable?

  • 6
    $\begingroup$ It has taken us a long time of a highly iterative process to get to the taxonomy system we have today on Earth. That did not detract from its value early on; it simply meant that refinements turned out to be required over time. I imagine that xenotaxonomy would be no different. $\endgroup$
    – user
    Commented Sep 21, 2015 at 14:49
  • 1
    $\begingroup$ Actually our biological classification system predates Darwin. It was created by Carl Linnaeus over 100 years before Darwin's book. en.wikipedia.org/wiki/Carl_Linnaeus $\endgroup$
    – Neil W
    Commented Sep 22, 2015 at 9:00
  • $\begingroup$ @NeilW Our modern system has been evolving for some time. Darwin introduced the evolutionary model, as I stated. I never said he invented the taxonomical system. $\endgroup$
    – Frostfyre
    Commented Sep 22, 2015 at 12:04
  • $\begingroup$ Darwin did not introduce the concept of evolution -- evolution is an observed phenomenon, first noticed and used by geologists. Darwin introduced a theory of evolution, that is, an explanation of how evolution works; it was not the first theory of evolution, it was not the last, and it is not the current theory of evolution (that would be the Modern Synthesis of Mayr, Dobzhansky et al.). $\endgroup$
    – AlexP
    Commented Mar 28, 2017 at 15:49
  • $\begingroup$ @AlexP I never said he introduced the concept of evolution. $\endgroup$
    – Frostfyre
    Commented Mar 28, 2017 at 16:02

6 Answers 6


Stable? Impossible.

Large changes to Earth's taxonomy have changed even in the last 20 years. As we get more into using DNA to map things more changes happen. We've discovered that things that look alike are not always closely related (they just both hit upon a successful design) and just because they are very different doesn't mean they can't be related.

So, with our current system we would try to start with large groups that make sense 'now' and we would have to be willing to modify any level in the future should actual life prove to not fit what we start with.

We wouldn't really need to have just 'Earth' and 'Extra-Earth' labels. I think having the location would become important too.

  • System: Sol
  • Planet: Earth
  • Kingdom: Animalia
  • Phylum: Chordata
  • Class: Mammalia
  • Order: Primates
  • Suborder: Haplorhini
  • Family: Hominidae
  • Genus: Homo
  • Species: H. sapiens

We, of course, would try to compare new life to what we know, but it will still be different and starting this way would keep things in better order, so even if we put something in 'animalia' on another planet it will still be preferaced by WHERE this animal was discovered. Since the chances are (unless interferred with by an outside visitor) all the life on the planet will be much closer related to each other than they would be to us or anything on Earth.

  • $\begingroup$ adding system/planet to classification might not be a good idea. what if you find same creatures on different planets? $\endgroup$
    – user902383
    Commented Sep 22, 2015 at 8:26
  • $\begingroup$ @user902383 I thought about that, The most likely cause for that is an intelligence that moved it there. On top of that, they will immediately start to diverge from each other I also think that you could hyphenate the system/planet to follow it's path, should you be able to figure it out. $\endgroup$
    – bowlturner
    Commented Sep 22, 2015 at 10:24
  • $\begingroup$ @user902383 - then the species would be categorised by the planet on which it originated. The chances of the same organism evolving on two different planets are so tiny that it can be safely discounted. $\endgroup$ Commented Sep 22, 2015 at 10:25
  • $\begingroup$ @GeoffAtkins yes you are right, but you might encounter same organism on multiple planets, but don't know which one is actual home planet. $\endgroup$
    – user902383
    Commented Sep 22, 2015 at 11:37
  • $\begingroup$ @user902383 - You would be able to (eventually) deduce the origin based on its similarities with other flora on that planet. For instance all life on earth shares genetic similarities; those similarities are likely to exist on other planets as well. $\endgroup$ Commented Sep 22, 2015 at 11:42

There are two options for categorizing life forms found on other planets/moons etc.

Method 1

Categorize life forms with respect to their location in the universe. For example we could have a naming system where creatures were named a, b, c, d etc after their planet name. So we could have Kepler-442b-a1 (species a1 living on Kepler-442b planet). This would be very suitable method of identifying where that life belongs to. Suitable for space scientists.

Method 2

Categorize life forms with respect to their chemical composition and respiration type. So we could have a CHO-Fe-O creature (body based on hydrocarbon and oxygen. uses iron oxidation respiration system). This naming system is more comfortable for biologists who are more interested in composition of organisms instead of where they belong to.

  • $\begingroup$ Your method 1 is going to become very unwieldy very quickly. For example, there are 350,000 species of beetle on Earth, so dung beetle could be Sol-3-217530. All in all, there are 8.7 million species on Earth. $\endgroup$
    – Frostfyre
    Commented Sep 21, 2015 at 16:24
  • $\begingroup$ You know, humans have a very nice ability to be able to classify things accordingly. So if there are species physically related to each other, a code can be used for their group => Sol-293b-btc223 = btc type creature, species index 223 on planet Sol-293b $\endgroup$ Commented Sep 21, 2015 at 16:32
  • $\begingroup$ For method 1, why not just prepend the solar system/planet to a meaningful species name? Earth Homo Sapiens Sapiens makes more sense than a long alphanumeric string. $\endgroup$
    – ckersch
    Commented Sep 21, 2015 at 22:02
  • $\begingroup$ Talk about naming some 30 planets with distinct meaningful names and then naming some 3000 or so distinct species living on each planet. I don't think the meaningful names are going to last long like that. Plus, scientific codenames tend to show some information about the object. So while method 1 and 2 show some information about the creature, simply naming them like earthly creatures doesn't really tell what they are. For example when we have 40,000 or so species discovered on different planets saying Blue-Ring Alpha noxus doesn't really tell where Blue-Ring is location and whats Alpha noxus $\endgroup$ Commented Sep 22, 2015 at 3:11
  • $\begingroup$ I don't think this would work in all sciences, but it does remind me highly of the categorized neurons of C. elegans, a nematode. The hemaphrodites of C. elegans have exactly 302 neurons. Their neural structure is so regular that a scientist can name them uniquely, such as V5.appap. It also does remind me of how we categorize influenza viruses, such as H1N1 $\endgroup$
    – Cort Ammon
    Commented Mar 28, 2017 at 16:32

Use a taxonomic system similar to what scientists used in the 1800s and then advance it as research progresses. As the OP states, it takes years/decades/centuries to work out a "correct" taxonomic system. Science on Earth went through at least three evolutions before settling on the modern one.

Early systems were based on phenotypical similarities that could be observed with the naked eye. Each new observation technology has shaped and altered our understanding of the animal taxonomy. First, high quality optics allowed closer inspection of animal morphology. More recently, DNA sequencing has added another insight into how animals relate to each other.


A xenotaxonomy for a given world will incorporate all the meta-taxonomy that we have derived from Earth's taxonomic structure. We know about evolution and how it works. We know about sexual and environmental selection pressures. So, while the new biosphere didn't evolve similar to how our biosphere evolved (that'd be really crazy if it did follow the same general track), we do know that it did evolve and we can use that as a starting point to identify and classify species and families.


It should be completely independent from Earth's life taxonomic. Specially if their biochemistry is very different (like using ammonia instead of water or silicon instead of carbon, different amminoacids, not based on DNA and RNA, etc).

In fact, how would you classify a jelly photosynthesizing three-headed and ten-legged creature living in ammonia's oceans that features spider-shaped cells with a sulfur-reducing organelles containing TNA capable of migrating in and out cells? Is this an animal? A plant? A fungus? A bacteria? No, it is something completely unrelated and different to what we have in Earth, so our taxonomy is completely worthless to them.

If the biochemistry is similar's to Earth one. You could at most have some creatures classifiable as weird and uncommon types of bacteria and viruses, but any earthly taxonomic classification beyond that would likely be completely invalid and unsuitable for those alien beings.


We need to consider:

  • the entire range of evolution, including possible 'post-singularity' entities that direct their own evolution ('provolve'). The taxonomy of 'provolvers' looks very different to 'evolver's, but perhaps when all provolvers are evaluated, patterns emerge by which to classify them. In any case, a single unified system would need to cover both evolvers and pro-volvers.
  • the basic physical form of the entity in terms of the phase of matter that it utilises. Most life as we think of it uses the solid/liquid/gaseous form. Possibly life forms could be based on plasma, bose einstein condensates or pure energy. All have featured in science fiction.
  • a distinction between life that evolved and life that was engineered. An engineered life form could be left to go its own way then evolve and/or provolve. By definition machine-life would fall in this category but so could engineered biological (or plasmoid etc) life.

Our terrestrial taxonomy appears as a tree because there are as yet no artificial life forms or provolvers. However a universal taxonomy would be a taxonomic phase space defined by a whole bunch of dimensions amongst the most basic of which could be those given above. A taxonomic group is then defined as all species starting within a given volume I (initial) of the phase space, passing through the set of volumes P{} via any pathway and having their evolution/provolution cease ((pro|e)volution ceasing would need to be defined) within volume E (end).

Based on the above classification terrestrial life would be seen to have explored a vanishingly tiny fraction of the possibilities defined by the phase space.

Having defined a phase space, deploying the tools and terminology of thermodynamic, infodynamics and entropy would then be likely to provide additional insights.

We should also be mindful that over the timescales represented the universe is co-evolving quite quickly so this provides another important aspect to roll into the whole framework.

Don't ask me to draw a diagram!

  • $\begingroup$ Since the concept of phase space is new to me (thanks!), I want to make sure I understand this. The initial volume consists of a certain set of parameters (say, has legs and must breathe) and the end volume consists of another set of parameters (say, has legs and has wings), so the taxonomic groups consists of all creatures that, for some portion of their evolution, began in the initial volume and ended in the end volume. Is that what you have in mind? $\endgroup$
    – Frostfyre
    Commented Sep 21, 2015 at 17:13
  • $\begingroup$ Wikpedia: "In mathematics and physics, a phase space of a dynamical system is a space in which all possible states of a system are represented, with each possible state of the system corresponding to one unique point in the phase space". Its important to add there is nothing to stop a system from returning to a prior point in the phase space as it evolves, even though it may be unlikely. And yes you have my intention correct (although it is just an example - there could be better ways of classifying similar pathways through the space). $\endgroup$
    – rumguff
    Commented Sep 21, 2015 at 17:27
  • 1
    $\begingroup$ I went over to Wikipedia to figure out what it was and wanted to make sure I was applying it here correctly. Also, if you wouldn't mind drawing a diagram, I'm sure future readers would much appreciate it. ;) $\endgroup$
    – Frostfyre
    Commented Sep 21, 2015 at 17:31

Assuming there actually is life worth categorizing in other places...

The problem with the location-based systems is that you're also not including any species that have relocated, or have been relocated. Humans that colonize other systems may result in different evolutionary paths (and recombinations). Same for other older intelligent, space-faring species we encounter. Worse for their equivalent of grain, apples, roses, chickens, cardinals, dogs and cats. That's also discounting UFO DNA complexes, panspermia, etc - which may really muddy the water.

Engineered life forms are a problem. As are combination life-forms.

When/where/how do you decide that life excludes all metal / AI? Or are robotic self-reproducing systems also needed to be organized under this scheme?

A different taxonomy for each original (if it can be determined) evolutionary niche seems plausible, with caveats for things thrown in later, and maybe another layer based on type (eg: uses 2-strand 4-base DNA).

We'll probably run fast gene samplers, and morphological search engines to classify new stuff that we encounter, on Earth and beyond. We may even get into cultural diffusion/memetic analysis, as lifeforms learn new methods/behaviors of coping with problems.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .