Over the last few years, one of the more interesting research topics in Uralic historical phonology and etymological phonology has been the Finnic sound change *ä-ä > *a-ə. Not only does this turn out to explain numerous other puzzles in Finnic etymology and historical phonology, but it also seems to be still accruing new evidence in its support at a surprizingly high rate. The amount of examples recognized in papers surveying the state of the research makes a good illustration of the ongoing frenzy of discovery, I think (transcription mine, in all examples):
- No more than two examples, *sappi < *säppä ‘gall’ and *talvi < *tälwä ‘winter’, have been consistently recognized across the 20th century; and they indeed go already way back to the 19th.
- Sammallahti (1988) still only accepted two other instances (and the latter has by now again fallen out of favor): *parci ‘beam, board’ < *pärtä ‘board’, *vaski ‘bronze’ < *wäśkä ‘metal’.
- Kallio (2012), working with only one additional example discovered a few years earlier: *kasi < *käsä ‘moisture’, was already able to suggest that the development is perhaps regular rather than exceptional.
- In parallel, Aikio (2012) expands the number by four additional examples, mainly words showing the later development *a > *oo: *koolë- < *kalə- < *kälä- ‘to wade’, *loomi < *lamə < *lämä ‘scab’, *pooli < *palə < *pälä ‘half’, possibly also *jarvi < *jäwrä ‘lake’.
- By Zhivlov (2014)’s defense of the development as a regular sound law, the list has grown again by two: *saksi < *säksä ‘dirt’, *soori < *sarə < *särä ‘vein, root’.
- The re-treatment by Aikio (to appear) list three more examples yet: *ahtëra < *äkštärä ‘barren’; *koi (< *kooji < *kajə) < *käjä ‘moth’, *sarni < *särńä ‘ash tree’.
4 to 14 (13 if we discard *vaski) is quite the growth rate in just half a decade. Compared with an entire century of stagnation before this, it should be evident that comparative Uralic studies have been earlier held back quite a bit by old dogmas like “Finnish should be presumed conservative by default”. OK, to be fair, the evidence is also swamped by later loanwords, of e.g. Germanic, Samic and probably also substrate origin, which are still being sorted out… So perhaps the most important methodological point behind this discovery has been the prioritization of the oldest, most widespread Uralic vocabulary over comparisons between neighboring language groups. (I often wonder how much e.g. Indo-European studies could also benefit from this same point.)
Moreover, the dust still hasn’t settled down. I am aware of at least a few recent individual proposals that do not appear to have made it into any published secondary sources so far:
- Heikkilä (2014: 86) adds *hanhi < *šänšä ‘goose’, a known Baltic loanword that can now be regularly connected with Erzya шенже /šenže/ ‘duck’.
- Kallio in fall 2014, on his lecture series on Proto-Finnic at University of Helsinki, has proposed *Soomi ‘Finland’ < *Samə < *Sämä. The word thus would be cognate with Sámi < *Sāmē after all, as has been long suspected — and in a much more direct way than earlier proposals (that typically appealed to complicated back-and-forth loaning thru Baltic) have suspected.
There might be other such cases out there too. And there definitely will be more to come: I have undertaken a more detailed hunt for examples, and have by now gathered no fewer than 11-12 additional examples to eventually report (including *sais- < *sańśə- < *säńśä- ‘to stand’, as covered yesterday on this blog). This will probably be more than enough for an entire paper somewhere down the line, but for initial release: watch this site.
This sound change seems also both distinctive and important enough that it could use some kind of a proper name. Traditionally Uralic studies have for some reason largely avoided naming sound laws, but perhaps this is because almost no developments significant enough to deserve a name have been even recognized so far? Simple (near-)unconditional shifts like *e > *i or *-k- > *-g- > *-ɣ- can be easily enough referred to by general phonetic terminology after all: “raising”, “medial voicing”, etc.
*ä-backing is new enough though that there are no clear options yet. While the change’s existence has been known for ages, there are no “classic” authors this could be attributed to, in the way I’ve opted to do with Lehtinen’s law. As for the present-day researchers who have contributed to its analysis, I do not intend on taking sides on who should gain top billing here, nor would I want to deprieve any of them of proposing an eponymous name for some other future or past discovery.
One possibility that still comes to mind might be “winter’s law”, after what is probably the best-known example (and, of course, in reference to Winter’s law in Balto-Slavic). But maybe that’s more punny than is proper for scientific nomenclature (outside of obscure invertebrate species, anyway). :) Liable to confusion too, perhaps. This type of approach could surely be further explored, to produce something like “gall/winter law”… But for now I will be going with the descriptive if less snappy “*ä-backing”.
[Edit 2016-03-06: of course, that needs to be “backing”, not “fronting”. Fixed now.]
See Bibliography for Aikio (2012), Sammallahti (1988), Zhivlov (2014).
- Aikio, Ante (to appear): The Finnic ‘secondary e-stems’ and Proto-Uralic vocalism. In: SUSA 96.
- Heikkilä, Mikko (2014): Bidrag til Fennoskandiens språkliga förhistoria i tid och rum. PhD thesis: University of Helsinki.
- Kallio, Petri (2012): The non-initial-syllable vowel reductions from Proto-Uralic to Proto-Finnic. In: SUST 264.