I'm skeptical of the 'secret tongue', it's more likely that practitioners of traditional medicine have their own specialized vocabularies, much like western doctors do.

IIRC from when I was studying Aymara, traditional medical practitioners had their own specialized vocabulary that was largely borrowed from another language (I want to say Uru, but that's a guess).

Yes, that does sound more plausible.

Well, I don't know; it was only a bit over a century ago that anglophone doctors stopped using Latin (as opposed to Latinate words in English) as a "secret tongue" to conceal things from their patients. So it might be more than a professional jargon, something more like thieves' cant or even Boontling.

Does anyone keep count of new languages appearing?

Well one problem is that languages appear at very slow rates (and their appearance usually isn't noticed until they've been around a while, a few decades at least).

They can die much more quickly though. Raise one generation without the language and it's almost for sure to die (though that can take a while too).

The 'secret language' is probably likely to be as Michael described. I wouldn't put it past Harrison et al. to draw conclusions from very minimal research and data collection, and very little, if any, discussion with linguists, anthropologists and ethnographers working with the languages he mentions.

This is what happened with the Living Tongues website's claims regarding some of the Bardi languages. Claire Bowern has written about it here, the discussion for which includes a detailed comment by Harrison himself. He maintains that it was not designed as a language documentation fieldtrip but as a journalistic exercise, to raise the profile of language endangerment generally. See also Jane Simpson's post for a more specific discussion of fly-in-fly-out (FiFo) fieldwork.

Ouch. I'm adding that to the post. Thanks!

On language creation: as I reported for the NYT two years ago in a story on the latest edition of Ethnologue, one way that new languages appear is because Ethnologue editors decide that two dialects, once collected under one language name, are actually two languages. That's why the total number of languages in the world grew from the 14th edition (in 2000) to the 15th in 2005. The editors explained in the foreword that they added 103 new languages in 2005.

There's a number of reasons for this recategorization: local political factors, more complete or more accessible data, or even language change itself. It's not the sort of de novo language creation of something like Klingon. And "newness" here obviously doesn't refer to the health of those languages. And of course it follows the way that Western science is given credit for "discoveries" that were always known by local people -- so I like Michael Farris's comment, that the knowledge of "newness" follows on actual newness by quite some time.

Which raises another point about this project: The figure of 153 appears in every article on this that deals even in part with the Northern Australia 'hot-spot', that is, the number of languages at risk of 'dying'.

I don't think the number is less, in fact I think it may even be more than that, either way, I'm curious as to how they were able to conclude such a precise figure, especially since:
1. The definition of 'language' versus 'dialect' is particularly problematic in Australia, with dialect chains comprising one 'language', even though the terminal languages in the chains are mutually unintelligible and therefore 'languages' with respect to each other.
2. The 'hot-spot' only covers a portion of the country, yet there are indigenous languages literally from coast to coast. Did they only consider the languages from within the hot-spot? Others are also desperately endangered.
3. Doing an exhaustive survey of languages in any linguistically-dense area is always going to be difficult and I'd hazard a guess that this is more the case in Australia. Moreover, it must have been terribly difficult to evaluate the status of languages without doing much background research, as Claire pointed out. And:
4. They didn't even go to some very major areas in the 'hot-spot', like Ngukurr, as Wamut points out (on the comment thread in Claire's post), even though they were scheduled to do so. Other areas, such as Kybrook, knew absolutely nothing of the project, even though the region surrounding Kybrook farm (which, by the way, is the focal township for almost 100 Kms in any direction) comprises some 6 or so languages, all but one in seriously grave danger (one, Jawoyn, 'died' in the last two months). Did they guess as to the status of these languages or did they rely on published material. If the latter, did they acknowledge their sources? They certainly didn't seek much information from people working in these areas.

I honestly think the number 153 is plucked out of the air. It's probably more or less on target, but wouldn't saying 'between 150 and 200' be a better representation of their methodology? Also, giving a specific number like 153 gives the impression that they have done an exhaustive survey and therefore, that we are much closer to documenting all these languages thanks to NatGeo's efforts. This is not the case.

Kallawaya: Regarding our 2007 field trip to the Kallawaya, we did indeed have a limited first visit. We are currently setting up a joint project to continue this work with a Linguist at the Bolivian Ministry of Education and an Anthropologist at the National Museum (who accompanied us on our trip).

Prior to our trip, and in preparation for it, we read the available literature on Kallawaya (whence the term "lengua secreta"), some of which I listed here for interested parties: http://www.swarthmore.edu/SocSci/dharris2/Kallawaya_Resources.php

One of our tentative findings was that although Kallawaya has been described almost exclusively in the literature as a "secret language" used primarily for medicinal purposes, it can also be used to express a very wide variety of everyday utterances. It thus resembles a full-fledged language more than a specialist jargon or cryptolect.

We presented our findings at the Ministry of Culture in La Paz, and archived our recordings with the National Museum. We will continue this as a longer-term project in the future and are currently planning our next field trip to Bolivia.