the core concept--classifying groups according to shared traits for purpose of analysis--is certainly not a problem.  you can classify anything according to any criteria you want.  i don't think anyone in the thread has disputed this point.  

the difficulty is developing robust criteria that can actually differentiate things in the way that we want them to.  if phylogeny is the basis for our taxonomy, the problem is creating categories with the minimum possible overlap, such that, say, convergently evolving taxa are not deceptively lumped into the same category.  or the other way around--by folk taxonomy, obama is black, even though that is a minority of his ancestry.  i have a dear friend who is half sri lankan and half german.  her mother is a goddamn aryan poster girl, but to look at my friend, she's completely south asian.  so do we just create a new file folder for every single individual?  this seems to erode generalizability, which is the entire basis for taxonomy in the first place.

and as i've posted numerous times, the problem is treating the abstract system of classification as something that inheres to actual human beings.  whether or not you agree with me on my overall point, you have to admit that this is a persistent problem with 'race'.

... And you do realize that the same criterion (cranial features) were one of the foundations of the antique research you are bashing, right?

There's no question that erroneous conclusions were drawn in the past, but if you wish to bash the research of old, do so on its legitimate faults.

You cannot on one hand accept the criterion used to identify Neandertals and other varieties of human, and at the same time reject the exact same criterion used in other research.

Height is a single phenotype that can require very little shared genetic history, or perhaps none at all. Race involves a broad collection of shared genes.

Founder effect is only one aspect of genetic differentiation of populations. There's also genetic drift, and selection pressure based on both environment and social organization. It's probable that hunter-gatherer societies are selecting for different personality criteria than agricultural societies. Sending aggressive criminals to the gallows for a few centuries in an agricultural society because they disturb the peace and make problems has probably changed those populations genetic makeup, while aggressive, nervy guys are probably more valued in hunter-gatherer societies. And, likewise, the genetic makeup of the population is probably affecting the society--the population and society are co-evolving.

i'm confused here--what exactly is it that you think i'm bashing?

great--what genetic expressions are to be included in the collection?  IOW, what are the specific criteria by which we establish race?

The research on racial characteristics and their geographic distribution in the last century.

You have implied (or outright stated) repeatedly that the criterion used in that field was baseless or meaningless.

Yet you readily accept one of those same criterion (cranial features) when applied to a different subject of study (humans of 40,000 years ago, rather than this century).

It would be accurate to view the 20th century research on the subject as flawed. But to suggest every criterion they employed was useless in the study of human variety and lineage is silly, and wrong.

second post winner!

J-

There's only one breed of dog, too. It's true, trust me.

ok, then you weren't tracking earlier.  i don't object at all to morphology as a tool of classification--general ancestry of a.m. humans can often be determined from cranial analysis.  and i completely acknowledge that many morphological traits are the result of geographical selection pressure.  no problems there.  and these traits often occur in fluid but somewhat regular constellations, so that's fine too.

what i object to is a rigid typological/categorical distinction between populations using traits that are shared among those populations...especially when the geographical isolation has largely been broken down.  i'll restate the example i used before: the traditional conception of "race" divides people into a hard-copy filing cabinet based on macrostructure, which can be deceptive.  as much as we try to redefine it in order to add nuance, the term is always going to carry the connotation of essential categorical differences.  for most people, obama is always going to be race = negro, regardless of his very mixed ancestry.

i find this to be a persistent problem.

So your science boils down to concerns about society and President Obama.

Got it.

if that's what you took away from that post, i guess there isn't much to discuss.

Opps read "species" instead of "race."

J-

Please explain how your social concerns fit into the scientific method.

Until you have done so, or demonstrate a separation between your personal bias and actual scientific analysis, why should your commentary carry weight?

my concerns are not social--they're terminological and methodological.  play "spot the morphology" on obama's cranium, and the result is african (based on the standard ID criteria).  the problem is that obama is half-nonafrican.  so clearly there's a methodological problem somewhere.  

and as i posted earlier, there's a reason that science doesn't recast the term miasma (even though it can be construed as relevant), nor are personality types still classified into choleric and sanguine (even though these categories are intuitively useful).

these concepts are not difficult to understand.  the fact that you saw the word 'obama' in a race conversation and immediately assumed it was a social point says more about your bias than mine.

OK, with the disclaimer I still think it is not a useful exercise to define groups below species in humans and that traditional "races" are social constructs, there are some broad misconceptions here. Taxonomy is at its heart a practical science, not an exact one.

First an isolated group of tall individuals has no expectation of getting taller - either a height or range of heights will be selected for, or they will drift (rapidly in small groups) taller or shorter with equal probability.

Single characters are never useful. An individual character is not used ever - on their own they will likely exhibit homoplasy (convergence or reversals) or may be pleisiomorphic (unchanged from ancestors). When suites of characters are used, synapomorphies which show true relationship and are more likely in general overwhelm homoplasy and show true relationships. Synapomorphies are homologous characters that show changes from the ancestral state that are inherited by the group of decendants produced by the ancestor that had the change (etomologically it means shared derived characters). This can be done with any character set from morph, anatomy, physiology, or DNA. For DNA, simply each position in an aligned set of sequences is a homologous character, and the bases are the states. Each position may exhibit any of the above, but by using parsimony (shortest tree length) or other approaches like Bayesian (spelling?) synapomorphies overwhelm homoplasy. Regardless of what type of character set, choices are made such that the right depth is resolved. Sequences that change rapidly resolve terminal nodes because they generate enough synapomorphies. For deeper nodes, the longer time causes synapomporhies to reverse becoming homoplasy and eventually they become un-alignable and the homology breaks down. Using physical characters the same applies - presence or absence of flowers isn't going to help resolve genera of flowering plants. That is the systematics of it, and today the DNA is the best roadmap by far and away. The DNA isn't "some other class of feature that is less important". It is the most fundamental and inherent aspect of the lineage because it dictates completely the range of phenotypic variation and sets of characters possible. Even though the DNA examined itself isn't the DNA that codes for the particular characters necessarily, it resolves the relationships.

Enter the taxonomist. No taxonomic revisions are being accepted without a phylogeny anymore. It won't get published. You take the phylogeny and someone makes cutoffs and makes the call. Always except in this group. There are many practical considerations - what is useful for identification, what is useful for putting reasonable amounts of variation in containers, what is useful for putting a reasonable # of subspecies in a species or species in genera, etc. Even more practica considerations are coded in the rules - does it cause renaming of a combination that is popularly known and heavily utilized like a wild edible? Again, look at the top level taxonomy - the systemetists know the protists are all over the place and so do the taxonomists - it just isn't practical to place the rest of the groups that are as deep or deeper than the named kingdoms into separate kingdoms.

It just proceeds. Someone makes the cuts. After that, they map out the characters on the phylogeny one at a time and assess where the characters are convergent or not. Synapomorphic characters are now obvious because the topology hasn't been influenced by homoplasy and is dictated by synapomorphies. If you have enough characters you will be able to use them in combinations i.e. like a key. One key difference at the end between two close groups may not be unique or unshared with taxa in other outer groups - you only compare that character if they make the other cuts. This gets published. If you disagree, you publish a dissent with re-anaylsis or based on your own new data. People either accept it or not. Eventually the people who are actually using it coalesce on a standard or they keep using their own versions separately. The ultimate litmus is "what is practical" not "what is perfect".

Some traits are genetically dominant, especially in cases where a complete cluster of genes is required to express certain features.

You should know that.


Admit nothing, deny everything and make counter-accusations. Amirite?

You brought your personal social concerns into a topic that should be about science. No matter how you attempt repackage that, you have shown that your personal bias is key to the framework of your argument.

Are you talking to me, or mcgredo?