zompist bboard

So, most of the languages of the world have at least two contrasting series of stops, most commonly based on a voiced/voiceless contrast or an aspirated/unaspirated contrast (or some combination thereof, as in English). There are some, however, that only have one type of stop; the Utian (Costanoan and Miwok) languages stand out in this regard, as does Spanish if you analyze all of the voiced stops as underlyingly fricatives (which WALS apparently does). I was curious if people here knew much about any other languages with only one phonemic stop series, and particularly about allophonic variation in those languages. For instance, how common is intervocalic voicing in languages that are analyzed as only having voiceless stops?

Check Nort's list of languages with small consonant inventories. I believe many there, such as Rotokas (the smallest) have only one stop series.

KathAveara wrote:Check Nort's list of languages with small consonant inventories. I believe many there, such as Rotokas (the smallest) have only one stop series.

Rotokas apparently has phonemic /p b t d k g/ though

What about Inuktitut? Or (varieties of) Finnish?

Rotokas has two though /b d/ are usually . only one stop series is less common there than you'd think, but there are still plenty of examples.

most Australian languages have only one stop series. also Chukchi, Uralic, and I'd look in North America. Eskimo-Aleut obv, and the ones in my list

my impression is that langs with only one plosive series tend to either be australian/dravidian or have voiced fricatives/approximants in their place, but I don't know if that statistically holds and there are plenty of counterexamples

Nortaneous wrote:most Australian languages have only one stop series. also Chukchi, Uralic, and I'd look in North America. Eskimo-Aleut obv, and the ones in my list

my impression is that langs with only one plosive series tend to either be australian/dravidian or have voiced fricatives/approximants in their place, but I don't know if that statistically holds and there are plenty of counterexamples

Speaking to the OP's point about voicing alternation, I remember trying out vowel length allophonically conditioning the voicing of the following consonants; I got some feedback here that suggested that wasn't too unusual for Australian languages.

Nortaneous wrote: and I'd look in North America.

Algonquian languages are a good place to look for this, most of them have a single series of stops w/ intervocalic voicing. I know it's also very productive in Iroquoian languages and Crow as well.

Most varieties of Quechua have just one stop series. (But not Cuzco dialect, which has three.)

Solarius wrote:

KathAveara wrote:Check Nort's list of languages with small consonant inventories. I believe many there, such as Rotokas (the smallest) have only one stop series.

Rotokas apparently has phonemic /p b t d k g/ though

Ok, I remembered wrong.

Proto-Uralic is usually reconstructed with a single stop series (with some geminates that are analyzable as clusters), but most Uralic languages have a phonemic voicing distinction. The original situation is only retained in the westernmost Siberian languages: Mansi, Khanty, Forest Nenets and Northern Selkup. I believe systematic non-phonemic medial voicing is attested from Western Mansi, not sure about anything else.

Then there's a bundle of Finnic varieties where voiced consonants were first developed (via consonant gradation), then they were lenited away so the language again has only one stop series. This includes Ostrobothnian and Savonian Finnish; Northern Karelian; and arguably Estonian, though it has a 3-way distinction in length of stop consonants and you could perhaps insist that the short lax b d g (can be voiced in some dialects) are distinctively lax while (half-)long tense p(p) t(t) k(k) are the real voiceless tenues.

I think Lule Sami (and possibly Pite Sami as well?) might also qualify, if you were to analyze the preaspirated stops as /hP/ clusters.

A similar case of a family that originally distinguished no stop voicing but today mostly does is Dravidian, though Indic loanwords have been a major part of this expansion. At least Tamil still has no contrastive voice (only allophonic voicing of single intervocalic stops), though.

Less-known single-stop-series languages seem to also include (most of?) the Yuman family, e.g. Cocopah, Maricopa.

Polynesian is also an example though perhaps too obvious to note.

Thanks guys, good stuff! I don't mind obvious suggestions, either; I have a bad habit of focusing on the obscure and forgetting the obvious.

An addition I forgot to mention last time: Chuvash.

And if you squint a bit, pre-Soviet Mari also works; it has a phonetic voicing contrast but are allophones of what are better analyzed as /β ð ɣ/. (Modern Mari has Russified these to /v d g/.)