kechpaja wrote: > On Thu, Apr 12, 2018 at 08:55:09PM -0400, Herman Miller wrote: >> Well, there's a bit of a difference between the river metaphor and the >> family tree one that we're accustomed to, in that big rivers form from >> little streams coming together, while trees grow from the roots up and >> branch out from there. So if you're thinking about classification in a >> genetic sense, you're more likely to come up with the image of a tree >> branching out. > > Not necessarily, though. When that metaphor was first mentioned, I > assumed it was in reference to a river *delta*, where a single large > river splits up into a number of smaller bits before ending up in a > larger body of water. I didn't actually have a word for river delta yet, so I made one, "gremu" (also new: tsesta "small river, tributary"; emba "estuary"). Tirëlat speakers often do things in multiples of 12, so it makes sense to have categories that are roughly based on powers of 12. At the base level is "dajtu" which is more or less a species. Then you get the larger groups (pjak) which are named after bodies of water: synapjak "stream-category" (genus, around 12 species) tsestapjak "tributary-category" (family, around 12 genera) serripjak "river-category" (order, around 12 families) embapjak "estuary-category" (class, around 12 orders) lağipjak "bay-category" (phylum, around 12 classes) mundapjak "sea-category" (kingdom, around 12 phyla) "Rhak" is another word used for the largest categories. Although at first glance this seems a lot like the standard rank-based biological taxonomy, there's a couple of notable differences. For one, the categories are not necessarily nested in a cladistic way; a river may be known by more than one name in various regions as it flows from its sources to the ocean. (As it applies to birds, you can think of the "Bird Estuary" flowing into the "Dinosaur Estuary" before it reaches "Vertebrate Bay"). The other difference is that these categories are defined by size, not chronological order of divergence. Again using birds as examples, we've got something like this: Aves Paleognathae Struthioniformes (ostriches) Rheiformes (rheas) Casuariiformes (cassowaries, emus) Apterygiformes (kiwis) Tinamiformes (tinamous) Neognathae Galloanserae Anseriformes (ducks, geese, etc.) Galliformes (pheasants, etc.) Neoaves All other birds (lots of orders) With the Tirëlat system you get something like this: ┐ Bird Estuary ┌┐ Ratite River │┌ Ostrich Brook │┌┌ Rhea Brook ││┌ Cassowary Creek ││└ Kiwi Creek │└ Tinamou Stream │ (Bird Estuary continues) ┌┐ Fowl River │┌ Duck River │└ Chicken River └ (Bird Estuary continues) That is, even though it's a ranked system, there's no need for higher ranks if there aren't enough related species to group together. Streams can flow directly into an estuary. Ostriches, with only 2 species, barely even require a genus in this system (and would probably be considered as a single "dajtu" in Tirëlat, as that can include two or three related species if they're similar enough). There's no Tirëlat equivalent of "Struthionidae" or "Struthioniformes".