New Study Shows Mystery 'Hobbits' Not Humans Like Us (phys.org)

According to a study published on Monday, diminutive humans that died out on an Indonesian island some 15,000 years ago were not homo sapiens, but rather a different species. The Homo floresiensis, known as "hobbits" since they looked like small humans, were found to be a distinct species based on the layers in the specimens' skulls. This discovery could be the end of one of the most heated arguments in anthropology.

Comments

On the origin of "species"

[wonkey_monkey]

were found to be a distinct species based on

...the abstract and ill-defined concept of "species."

Re:"most heated arguments in anthropology"

[serviscope_minor]

The way high school students are taught about what a species is is simplistic; or at least it seems to be to me after years of working with scientists to support their data needs.

Absolutely, yes. The tree of life is not a tree. it's a directed acyclic (has to be causal) graph. At coarse scales, it looks very much a tree. Defining species is essentially the same as performing clustering on nodes of the tree (or leaves if you happen to be a cladist) such that the clusters represent a good compression of the graph in that the tree structure of the clusters is similar to the structure of the underlying graph.

Choosing the right cluster size and partitioning scheme, as with any clustering problem is very hard and the results will never be perfect. If the clustering is too coarse, it lacks descriptive power, and likewise if it's too fine, it over segments and the clusters aren't semantically meaningful.

Most of the time, fortunately, the clustering is easy. Humans for example, fir very neatly in to one cluster, and we look noticable different from and cannot breed with our pair of closest relatives.

The breeding thing covers most of it. Can breed -> same species, can't breed -> different species.

It's not perfect because some closely related species such as horses and donkeys can produce offspring ans sometimes those offspring are also fertile. However, they're not often fertile, the males never are and they have different numbers of chromosomes.

Then you get ring species ,where A can breed with B, B with C, C with D, but A can't breed with C and etc. A and D may or may not be able to breed depending on the type of ring.

And that's animals. Then you get plants which are all kinds of fucked up.