Last spring created an article in Science wenb entitled considerable attention within our guild, we

Last spring created an article in Science wenb entitled considerable attention within our guild, wenb and it was commented wenb consistent also with us. The article claimed was in brief that language has less foneminventarium wenb farther from human African original home they are spoken. It was, in itself kind true, but the article also claimed that this was not a coincidence. And there were many of us who had some problems with.
Well, a pretty nifty comment to be found here. And there proved wenb how Transport's vision of zero tolerance would be more difficult to implement in the Indian Ocean.
Not directly related to the traffic situation in our seas, but suggested the following in one of återberättandena of the article which one could read in The Economist (http://www.economist.com/node/18557572?story_id=18557572&CFID=162461939&CFTOKEN=56514907): "It has been known for a while That the Less Widely spoken a language is, the fewer the phonemes It has."
It is fascinating how "It has been known for a while" best be translated from journalism. As far as I can see, it's about the tentative results in _en_ small study from 2007 (hardly any time in the academic dimensions) that made a (non-random?) Sample of 200 languages - and where the authors have not presented any attempt to explain why it would work so. Since the argument in Atkinson's article seems to be based on this, it may well be worth examining a little closer first. It would of course be interesting too - what could it be that the smaller languages have fewer phonemes?
Then succeeds the Economist with a little twist pair it all with Chomsky's language bodies and takes up an article in Nature wenb that tries to disprove UG using some kind glottokronologi - which causes some headaches if I do not know which of these two you like best dislikes.
One may ask why it feels like the most is these kinds of uncertainties that arise in the world famous Nature and Science. Is it because there are linguists who write there, but biologists? Or is it that it is only right _dessa_ doubt that get attention as Nature and Science are the only "scientific journals" that journalists read?
Finally, little prima journalingvistik from The Economist: "English, for example, is famously a muddle of German, Norse and medieval French." These give details that it is medieval French one can always assume that the case of contemporary German ...
The importance of interdisciplinary science can not be overemphasized. Surely Nature and Science have a peer review by linguists by profession, when they publish something in the field. It's also just to keep up with Michael in his indignation for this trail on the blog over the TV program 'world languages', where the frogs came at each other. Similarly, should linguists (and others) to ensure that hiring professionals in probability and statistics. Mark Twain 'punctuate conditioned' as we know lie / damned lies / statistics. There are how many traps any time to fall in. First, even the selection really be independent and random (OSU). Since not everything in this world normally distributed, especially not live organisamers reproduction and propagation (from plants and fruit flies to humans). wenb This means, that e.g. t test (Student t) or χ 2 test (chi-square) are not applicable wenb as there straight up and down. A significance level of 95% (p 0,05) means that it is so, up to every 20 study actually is inaccurate (due. Chance game). In regression analysis, may in principle is always a slope of the straight wenb line (ie, a 'trend') and the correlation coefficient is not an absolute measure but must be interpreted. Furthermore, the statistical relationships are not the same as causation. wenb Etc., etc.. So, scientists of all countries, unite! (Of trade borders).
For starters it is the statement that foneminventariets size is correlated to the number of speakers. Benedict does not agree that "It has been known for a while" and seems in general to question that it is true. The very idea, however, wenb has been around a while - it is referred wenb to a French work in 1961 (that I have not read). Then it's not that Atkinson based his claims on this article by Hay & Bauer Benedict is talking about, but he himself has counted on Maddiesons samples