positive anymore

English, like most if not all languages, has what are called Negative Polarity Items (NPIs). NPIs are words or phrases that have to be scoped under some sort of negation, irrealis, or otherwise nonaffirmative clause. One example in English is "ever". We can say I haven't ever been to Atlantic City, because "ever" is scoped under negation. We can say I wonder if John has ever looked at syllable-initial geminates, because if-clauses are irrealis or nonaffirmative. We can ask Have you ever ridden an elephant?, because questions are nonaffirmative (they don't contain any at-issue assertions). But we can't say *I have ever been to Jane's house, because this is a declarative, positive sentence that makes an at-issue assertion.

For most(?) people, "anymore" is an NPI. Thus for most native English speakers, I don't smoke anymore is fine, whereas *Young people are so rude anymore is bad. However, there is a small subset of American English speakers (and possible speakers of other dialects) for whom "anymore" can be used in positive contexts, as in the second preceding example. My grandmother was one of these, which is probably the only reason I know this. For her, it was fine to say "The buttons on phones are so small anymore". I can't think of a good way to easily find good examples of these constructions on, e.g., google or COCA. Suggestions would be welcome.

might didn't

A few weeks ago I caught myself saying "might didn't". When it came out I was confused, and assumed, like my cohorts, that it was a speech error. But the next day I did a bit of google searching, and realized that in fact "might didn't" is a construction associated with the use of double modals, and that perhaps I was successfully integrating double modals into my grammar.

One example: "Great Tools For YouTube And Online Music Streaming You Might Didn't Know Of"

In standard English this would be "Great Tools...You Might Not Have Known Of", or hypercorrect "...Of Which You Might Not Have Known". The "might didn't" construction indicates to me that syntacticians are missing the boat if they claim double modals are merely lexicalied constructions inserted whole into T. "Did" isn't a modal, and there are clear syntactic parallels between "might could" and "might did". What we need is a theory of grammar for double modal dialects that correctly accounts for the pattern of usage, not a theory that best fits standard dialects and half-heartedly accounts for certain superficial aspects of double modal grammars. Personally, I'm interested in a proper syntactic account of double modals because I'm all for accurate description of minority languages and dialects, but I have a feeling that such dialects could also reveal important things about what might could be a part of Universal Grammar.

past participles

Many verbs in English have three basic forms: a present tense form, which consists of the bare stem (plus -s in 3rd person singular), a past tense form (usually -ed), and a past participle form (often irregular, but quite a few end with -en). Notice I say "many" and not even "most", much less "all". There are infamous examples of verbs like lie/lay/laid vs. lay/laid/lain, which have three distinct forms but overlap in two of them, or teach/taught/taught, which has an irregular past tense and identical past participial form. The participial form appears in perfect constructions such as "I had just gotten to work when the boss walked in" or passives like "The wine was drunk in less than an hour". However, since many such participles are rarely if ever used, some people are uncomfortable using some of them, or simply unaware that a separate form exists. (Test yourself: I have swum, or I have swam? Swum is the historical past participle.) While I am somewhat of a past participle enthusiast, I rarely really notice the substitution of the simple past with verbs like "swim" and "drink". The ones that do strike me as odd are those that I perceive as common, which is why constructions like "was began" catch my ear. "Begin" is significantly more common than "swim", as evidence by the 106,952 hits for "began" in COCA versus the 2,069 for "swam". Likewise, "begun" gets 19,007 hits while "swum" only gets 186. And note that the "begin" to "swim" ratio is twice as high in the participial form than in the past. I think it's for this reason that constructions like "was began" strike me as odder than mere "had swam". COCA gets 26 hits for "had began" versus 4,865 for "had begun", and 3 for "had swam" versus 59 for "had swum". In other words, the past-for-participle substitution rate for "swim" is an order of magnitude higher than for "begin" (.5% for "begin", 5% for "swim"). At this point in the evening I'm not about to embark on a frequency analysis journey, but my guess would be you'd find similar patterns for many other verbs: past-for-participle substitution rates rise as raw usage frequency decreases.

pre-fixed menu

I saw a nice example of an eggcorn the other day, advertising a "pre-fixed" menu. This is of course referring to the phenomenon of a "prix fixe" menu, where restaurants offer a multi-course meal for a set price (typically one that's cheaper than expected). Even though I don't speak French, I've always thought this phrase was fairly transparent: as an English-speaker, I'm familiar with the fact that French adjectives (like most other Romance languages) have adjectives after the noun -- we even have some traces of it in English, e.g., Attorney-General. And if you know that much, it's not a big stretch from prix fixe --> price fixed --> fixed price. But for someone who's only ever heard the phrase pronounced, the similarity might not be as obvious: /ˌpriːˈfɪks/. This certainly does sound almost identical to a standard pronunciation of "pre-fixed". And since prix fixe menus have a price that's already set, the semantic notion of a menu being "pre-fixed" makes sense as well. Phonetic similarity + semantic compositionality = the perfect scenario for eggcorn formation.

Just for fun, some examples that I found online:
http://www.afridom.net/lesouk-harem/menu.html
http://www.thespot-restaurant.com/party.html
http://www.clarksbarandgrill.com/Menu/Dockside-Grill-pre-Fixed-Menu.htm

N N N N N N

Just a quick post before I get back to reading Seth Cable's recent article on Tlingit and Q particles. Last month CNN ran an article called "Titanic 100th cruises spark buzz, debate". Even before looking at the content of the article, I understood the basic gist of this headline: there are going to be cruises on the 100th anniversary of the Titanic's sinking, and these cruises are sparking buzz and debate. However, I find it amazing that I was easily able to understand what seems like it should have been a crash blossom. The headline is a string of six words that could almost all be nouns or verbs. Titanic and 100th can only be nouns, but the other four words could go either way. The anniversary is cruising some area called a spark buzz? Without the comma there would be been even more possible permutations.

Let's see if we can't do that

Last week Vicki Hartley wrote to ask me about "the use of a negative construction when a positive construction is simpler". One example is "I'll see if I can't do that". Strangely, this sentence means essentially the same thing as "I'll see if I can do that". Another example she asked about was, e.g., "I don't have but three students", to mean "I have only three students". The way I see it, these are two separate constructions, though there is an underlying theme.

In the first, you're saying that you'll try the opposite of what you intend. I told Vicki that I saw this as similar to what goes on in the scientific method -- we'll start with our hypothesis and try to disprove it; hopefully we'll fail. Logically, this seems to make sense, since either way the result is the same: you find out whether or not you're able to do something. On the other hand, as a speaker, I wonder if this "scientific method" approach is really what's going on here. To me "I'll see if I can't do that" really doesn't mean "I'm going to attempt to not do that". To me I can't find any semantic difference in "see if I can't" versus "see if I can". Pragmatically there are differences -- the redundant negative version seems to presuppose that there will be some difficulty associated with the course of action, hence the negative. On the other hand, it seems to also presuppose that the course of action will result in a positive outcome. To me, telling a sopping wet child "let's see if we can't find you some dry clothes" would be infelicitous if it were my child at my house. I could only felicitously say that to my child's friend at my house, where there would be no reason to expect to find them dry clothes that fit properly. (An aside: I picked this example because I associate "let's see if we can't..." with a parent talking to a child -- not sure if this is relevant to the discussion at hand.) However, I'll also say it implies that I'm relatively sure of a positive outcome (finding dry clothes that fit). If I say "let's see if we can find you some dry clothes" I don't get quite the same expectation; there's more tolerance for the negative outcome "oh well, I guess not".

The second construction is a slightly different version of the redundant negative. For me, "I don't have but three students" is slightly marked, but fully grammatical, whereas "I have but three students" sounds archaic almost to the point of ungrammaticality. So we have the following, which all entail having three students:

1) I have three students.
2) I have only three students.
3) I have but three students.
4) ?I have but only three students.
5) I don't have but three students.

I have a question mark by (4) because I'm not sure if I like it or not. I
think I don't, but it doesn't seem totally wrong. And then we have the
following, which all entail NOT having three students:

6) I don't have three students.
7) I don't have only three students.
8) *I don't have but only three students.

I'm pretty sure (8) is bad, but if it's not, I think it would mean the
positive, not the negative.

My guess is that we should treat this second construction as parallel to the first one. Despite my archaic interpretation of "I have but three students", this was definitely fine in earlier versions of English, and my guess is that many people would find it unremarkable even today. Thus "I don't have but three students" is essentially the same phenomenon as "Let's see if we can't do that". One avenue of research that might prove useful is semantic research on some of the North American languages. Salishan languages have suffixes that create a "managed to" reading (non-control transitivizers, for those in the know). Navajo has an adversative reading that indicates that a proposition is counter to expectation. Blackfoot (an unrelated language) has the same thing (which coincidentally is also the affix for "please"). This might be what we're seeing in English: variation based on presuppositions of the speaker's ability to bring into being some desired or discussed resultant state.

new Clarion blog post on writing systems

If you're interested in writing systems, check out my post this week for the Clarion Foundation blog: Writing Systems. As an amateur writer, I sometimes wax a little authorial on these posts, but if you've found them too fiction-oriented so far, know that I'm intending future posts to be more strictly linguistic in nature.

"the written word"

I was interested to see the headline "LOL -- 'Webspeak' invades Oxford dictionary" on CNN this week. The article is little more than a blurb about some new additions to the Oxford American Dictionary, but I was struck by the first line: "Are years of e-mails, text messaging and status updates finally affecting the written word?" When I read that I did a bit of a double-take, as you might be doing right now. "Hold on a sec," you might think, "aren't ALL of those things written words?" This usage takes to the extreme the idea that "the written word" as a set phrase is somehow not compositional; it doesn't literally mean "a corpus of written materials in contemporary usage", but rather some lofty edifice culled from esteemed writers and curmudgeonly literary critics. While I acknowledge that "the written word" is a semi-idiom in many dialects of American English, I would never use it quite as idiomatically as in this article -- literally juxtaposing a huge corpus of written material with the ethereal ideal of "the written word".

Despite this opening line, the article isn't critical at all of this move by the OAD. The author in fact notes that "It is nice to see Oxford attempting to get with the times" by including expressions that many of us see every day. Lexicographers are often remarkably descriptive, despite the tendency for prescription among those who use their products regularly. However, the author does fear that this will make difficult times for English teachers, as students back up their usage of TTYL and LMAO in academic writing with dictionary citations. I can certainly see English teachers cowering in terror, even though this seems to me ridiculous. As long as we talk about what is appropriate rather than correct, there's no need to fear descriptivism. For instance, I rail against those who teach that it is "incorrect" to use "which" in restrictive relative clauses, or that it is "ungrammatical" to use double modals. On the other hand, in some contexts there are reasons for teaching that it is inappropriate to use these in academic papers (although frankly I'm always against the claim that "which" should be only used for nonrestrictive relative clauses).

The problem with the absolutist view of English is that it isn't absolute. If you try to teach students that "which" should only be used for nonrestrictive relative clauses, you don't have anywhere to turn. The dictionary won't tell you this, esteemed authors don't show this usage, even the venerable old Strunk didn't keep his whichs and thats complementary (though when White came along he added the rule and edited all of Strunk's examples to make them fit the rule). Too often what people think of as "correct" grammar is simply bits and pieces of inconsistent jargon they've internalized from many different, often conflicting, sources. What students need to be taught is that academic writing is a formal style with strict rules. It's not that double modals are wrong, it's that double modals are frowned upon in academic writing. And that's a reason not to use them in such contexts if you want to get a job.

Compounds

English is mildly notorious for its non-compositional compounds. In this case I don't necessarily mean that compound words or phrases have nothing to do semantically with their components, but rather that the relation between the components is somewhat unstructured: there is no strict relation between X and Y for a compound X-Y. One relatively well-used example of this is the difference in the semantic relation between the two components in "olive oil" and "baby oil". You make olive oil by squeezing olives until the oil runs out of them. This is not how you make baby oil. In fact, the first word in "baby oil" has a completely different relation than the first word in "olive oil". In "olive oil" the first word indicates the source of the primary component, the oil. (Compounds in English and some other languages are right-headed, meaning that the component on the right gives you the category and basic sense of the compound: "olive oil" is a type of oil, not a type of olive.) In "baby oil", on the other hand, the first word tells you something about how the oil is intended to be used. You can see the same difference in "spring water" and "holy water". Holy water may in fact be spring water (I'm not sure if churches typically use bottled water, tap water, or some specially sourced water for this), but "holy water" indicates something different because it indicates what the water is going to be used for, rather than where it came from.

One thing I hadn't thought much about until recently is that it's not just N+N compounds that behave in this peculiar way. For instance, there's good reason to be afraid of baseball-sized hail, but no real reason to fear a family-sized bag of candy. Like the N+N example above, these types of adjectival compounds can refer to completely different types of relations. Hail that is baseball-sized is the size of a baseball, but a bag of candy that is family-size is not the size of a family; rather, it's a bag that is a size appropriate for a family. English is not the only language that has these types of unpredicatable compounds. Blackfoot also has some unpredictable compounds. One of my favorite of these is the word for horse, ponokáómitaa literally means elk-dog, where ponoka is 'elk' and ómitaa is a bound form of the root for 'dog'. Presumably this stems from the association of horses, when they were first encountered a few hundred years ago, with the general ungulate form and size of an elk, and with the beast-of-burden function of a dog, which the Blackfeet used to carry travois and other equipment.

double modals

Double modals are a feature of certain dialects of American English. For a long time I thought this construction was limited to the South, but I have since learned that there are other areas of the country featuring double modals, including Michigan. Some of the more common examples of double modals include "might could", as in "I might could do it" for "I might be able to do it", or "used to could", as in "I used to could do it" for "I used to be able to do it". Because these constructions are stigmatized in prescriptivist grammar (and perhaps because they don't occur in the Northeast or California), they haven't gotten a lot of attention in the linguistic literature. Given that current syntactic theory places modals in the T node, it's unclear how we should represent something like "might could".

One idea that non-speakers of the relevant dialects have is that double modals are redundant or unnecessary (in addition to of course being incorrect). However, to speakers of these dialects phrases like "might could" and "might be able to" are not in complementary distribution. In an example like "might should", the double modal indicates a very different information state than "should". A better translation would be "It might be the case that I should", expressing perhaps an irrealis deontic mood. A good example of non-speaker confusion can be found in one of the COCA hits for "might should", referring to a Southerner saying "he might should go" rather than "he might go". Of course, any person that speaks a double modal dialect knows that these two are not at all equivalent semantically or truth-conditionally. My point here is not to poke fun at those who don't know how to use double modals (although I might should poke fun at those who poke fun at those who use double modals). My point is simply that double modals enrich the English language, sometimes creating a shorter way to indicate an information state ("I might could" versus "I might be able to"), and sometimes conveying information that can't easily be conveyed any other way. It fills a niche in the same way that y'all or youse fills the want for a contrastive second person plural pronoun.

Crash blossoms

Recently the term "crash blossom" has come into use to mean a news headline so clipped and ambiguous that it becomes nearly impossible to get the correct reading the first time around. The term comes from the headline "Violinist Linked to JAL Crash Blossoms", which on first glance might garner the interpretation that some violin-playing individual has something to do with something called "crash blossoms" owned or manufactured by some company JAL. In actuality the store is about a plane crash victim blossoming into a superb musician. You can check out all sorts of other examples at crashblossoms.com and Language Log (e.g., here). The reason behind this is that as we read, we construct a possible syntactic structure in our head, and when you have long strings of words where each one can be either a noun or a verb, you runs into problems. For instance, in "crash blossoms", "crash" can be a noun or a verb; "blossoms" can likewise be a plural noun or a 3rd person singular verb form.

I came across a gem the other day: "Stabbings suspect an enigma". This isn't a crash blossom that's genuinely ambiguous if you semantically interpret every word as you go, but apparently I was reading a little too fast for myself when I looked over this, because all I really saw was "N-pl V.3rd NP", which would get interpreted as some stabbings suspecting the existence of an enigma. Obviously that's not the right interpretation. Really what you get is two NPs with an omitted copula. This is not the type of thing that would be ambiguous even for a second in spoken language, since the noun and verb forms of "suspect" have differing stress. But then this type of headline isn't the kind of thing that would be spoken at all, which is why we get some great garden path sentences from terse copy editors.

My computer

Over at Not Always Right one contributor recounted a humorous tech support tale about a customer who didn't understand the command "double-click on My Computer": "How can I click on your computer? You must be thousands of miles away!" Besides the obvious humor of misinterpretation (easily understood for the technologically illiterate, since the PC standard system icon "My Computer" co-opted a regularly occurring noun phrase), there's an interesting issue about prosody.

For me, "double-click on my computer" and "double-click on My Computer" are not pronounced identically. My guess is that this is true for the majority of English speakers, or at least native ones (feel free to submit dissenting opinions below). Because "my" is a functional morpheme indicating possession, it often ends up cliticized onto the noun it modifies. While many speakers have the full [maj] in careful speech, rapid speech will often produce simply unstressed [mə] in many dialects. In many languages this reduction has gone a step further, so that the possessive morpheme is now phonologically and morphologically bound to the noun in question. In my own speech, this destressing is realized by "my" getting only secondary stress, rather than primary stress, in a sentence like "double-click on my computer". On the other hand, when we're talking about "My Computer" qua shortcut to hard drive contents in Windows systems, the otherwise identical phrase takes on a new life. "My" is no longer just a clitic. In my own speech this is realized as primary stress on both "my" and the second syllable in "computer".

The presence of absence of destressing in certain words can thus give us clues to meaning and the parsing of certain phrases. For instance, if I'm on the phone with tech support and they tell me to double-click on jɔɹ mjuzɪk, I'm going to assume that there's some specific system folder I'm unaware of that's labeled "Your Music", whereas if they tell me to double-click on jɹ̩ mjuzɪk, I'm going to assume they want me to open the folder where I keep my music.

Ellipsis

Just north of my apartment there's a popular diner that has their own printed paper placemats. Printed on these are various slogans of the general form "Take time to X, it is the Y of Z", e.g., "Take time to THINK, it is the source of power." Leaving aside the somewhat obvious nonsensicalness of othese platitudes, they use an interesting form of ellipsis that caught my eye (or maybe my language faculty). Perhaps ellipsis isn't even the right word, because what I noticed is the odd reference of "it" in these cases. In "take time to think", "think" is a verb, whereas "it" in the second clause is a noun, and refers to a noun. So how can this be? Pragmatically there's no ambiguity, and I'm sure most people don't even notice that the construction's strange: obviously thinking or thought is what is intended to be "the source of power" (hence the old syllogism "Knowledge is power; power corrupts. Study hard; be evil."). Though it's fairly easy to parse, I can't even remember coming across such a construction before, where a pronoun refers back to something that's technically the wrong category. This seems somehow different from pragmatically instantiated referents, because the referent is in one way overtly present (the verb "think"), but in another way completely absent (there's no gerund or noun "thinking" or "thought").

posting on Clarion Foundation blog

I'll be posting occasionally on the Clarion Foundation blog on the linguistics of created languages in sci-fi and fantasy, so today I'll just give a link to my first post over there, an expansion of something I did a while ago on this blog: http://clarionfoundation.wordpress.com.

Feel free to leave comments here or there, including topics you'd like to see dealt with if this is the sort of thing you're interested in.

Language and place

I've been thinking recently about the connection between language and place, not coincidentally because that was the theme of the SILS conference I attended last month. As an English-speaking American of German and Scotch-Irish ancestry, I haven't in the course of my life placed a lot of weight on the relationship between a language and the geographic location where it is spoken. No doubt this is probably partly because of the omnipresence of English, even in countries where it isn't the native language. Certainly there are plenty of dialects, even just in America, but I was born in Florida to parents from New Jersey and California, and went to school in Tennessee and Montana, and I've been speaking the same English the whole time, noticing only slight differences in the English spoken to me in those different places. It is definitely the case that American dialects are fading in younger generations as people acquire more language from standardized sources such as television and the internet, and people move around more and are exposed to people from many different regions. But that's not what I want to talk about today.

I think the issue of language and place is a difficult one for Americans because 99.8% of us natively speak a language that was imported within the last 500 years (though in my own family I only have to go back 100 years to find someone born in Germany). On the other hand, Salish peoples have been speaking Salishan languages in British Columbia for around 10,000 years. While there have been changes in the language (10,000 years bp would probably find a single group of people speaking Proto-Salish, whereas today there are more than twenty mutually unintelligible Salishan languages) and migrations (modern Salishan languages stretch south and west in Montana), there is certainly some truth to saying that 10,000 years you would find the same people speaking the same language. This relation with place can be found in the languages spoken there. No, I'm not going to claim some Sapir-Whorfian causality between the geography of British Columbia and the Salishan languages. But the time depth of lexical items and semantic change in cognates can be used as a somewhat reliable method for investigating migrations. (For instance, the time depth of words for certain types of trees can be used to map some Indo-European migrations in and out of areas containing those tree species.)

For indigenous peoples, there is typically a strong, often spiritual, connection between language and place. Yes, every people is indigenous to somewhere, but I think the most appropriate use for the term "indigenous" is to mean people who live in the general geographic area of there ancestors. For instance, we could say that the Navajo are indigenous to America, since Na-Dene peoples have been here for tens of millenia, but not necessarily indigenous to the southwest, since Apachean peoples migrated there within the last millenium. Some people talk about the fricatives and labialized consonants of Northwestern languages recalling the lapping of waves on the shore. While I agree that's a beautiful image, it doesn't really have any objective fact to it. On the other hand, the fact that Tsimshianic languages have affixes meaning "upriver" and "downriver" does say something about where the languages are spoken. At the risk of touching on the language/culture debate, we wouldn't expect to find such affixes in a language that's been spoken in a riverless desert for 5000 years.

The slogan for the SILS meeting last month was "For every place a language". I like it. Because there is a language for every place, and I think it's something that Americans often forget. People are too quick to say "Learn English if you want to live in America", even though most of our ancestors didn't bother to learn Cherokee or Delaware or Kitsai or Karuk when they came to America. Perhaps we don't like to think about such things because we don't have those ties to our ancestors. My grandmother's grandfather lived all his life in a place I have never seen, and all his life spoke a language I can't understand. Maybe we devalue the connection between language and place because it's something most of us can never hope to have.

Tongue twisters

(I'm using Lucida Sans Unicode for the phonetic transcriptions in this post; I think most people have this on their computer, but if something's not rendering properly, you probably don't.)

As a phonologist, I'm always interested in tongue twisters. One of the classics in English is "She sells seashells by the seashore": ʃi sɛlz siʃɛlz baj ðə siʃɔɹ. Especially in the IPA version, it's easy to see the proliferation of alveolar and postalveolar fricatives, which is the source of difficulty in pronunciation. For English speakers, even simple words in other languages can be tongue twisters, especially if they contain sounds that aren't present in English, such as lateral fricatives or uvulars. The Kiksht word for 'eight' is a great example of this: ɢutɬqt. My recent facebook post about the Okanagan word for 'thistles' (sntkwlkwall'iw'stn' -- don't worry, there are some epenthetic schwas in there) led to some discussion of some of our favorite-sounding words in other languages. Bill Poser mentioned the Shuswap word for 'juniper': punllp (where ll is a lateral fricative), and I mentioned Bella Coola lhk'w-, 'tiny' (where hl is again a lateral fricative). Mithun (1999) gives a great one in Gitksan: nagáksdiː gáʔaɬ ɬagaχgáːkxʷɬ ɬagaxʷɢákʷɬ ɬagaχq’áːχɬ ɢáːqʰ, 'I have just seen for the very first time the toughness of the sinews of the wings of the raven.' Here the difficulty is in the combinations of lateral fricatives, velars, and uvulars, where velars and uvulars contrast within a syllable, and voicing varies. Adding to the difficulty is the similarity of the words. Also included is one from Choctaw: ʃɔ̃ʃi ʃwa ʃwakã iʃowã, 'Do you smell a stinking worm?' Besides the fact that people like to have fun with language, tongue twisters are probably common because people who speak disfluently are seen as less prestigious than those who are able to speak naturally and without error. Obviously nobody is capable of flawless speech 100% of the time, but we definitely judge those who make noticeable errors. Case in point: the Bushisms industry.

Steven Menefee 1981-2010

I just found out today that Steven Menefee has passed away. I only just learned that he had cancer last weekend, though I gather he had been fighting it for a while. I first met Steven at the Workshop for American Indigenous Languages at UC Santa Barbara in 2008. He and a couple colleagues from University of New Mexico had come to present on creating linguistics terminology in Navajo. After the talk I went up and introduced myself, told them how much I was interested in what they were doing, etc. That probably would have been the end of our contact, but Steven was such a friendly guy that when he saw that we were staying at the same hotel he shouted across the courtyard at the Super 8 to see if I and my University of Montana colleagues wanted to join them for dinner. Steven knew the area, so he was our guide to a rather wild night that summer weekend in 2008. The only other time I had the pleasure of his company was in Albuquerque at the High Desert Linguistics Conference that fall.

I saw his name on a presentation at SILS last weekend, so it was an unpleasant surprise that when I got there and asked his UNM friends if he was there, the answer was crestfallen faces and an explanation about Steven's cancer. Steven was one of the most caring, friendly people I have met, and a linguist who cared deeply about the people who spoke the languages he studied. He was a man of strong convictions and above all open, warm-hearted compassion. He laughed easily and always created a positive atmosphere with those around him. He struck me as the kind of person whose goal in life was to make the world a better place, and I can say without a doubt that it is indeed a better place because of him.

We'll miss you Steven.

In Oregon

I'm in Eugene, OR for the meetings of the Stabilizing Indigenous Languages Symposium, Athapaskan/Dene Languages Conference, International Conference on Salish and Neighboring Languages, and Hokan-Penutian Languages Conference. It's been a pretty good weekend, with some good presentations, and I've gotten to meet some fun and interesting people. I just got out of Leanne Hinton's keynote address, and soon it will be time for the conference dinner.

A minor note of displeasure has to do with my native language. My native language is English. I love the English language, and as much as I want to teach my children to speak another language, I know that it's not going to happen, both because I'm not fluent in any other language, and because English is a big part of my own heritage and that of my parents and grandparents, and I want my children to share in that heritage. I recognize that English speakers often exert an oppressive force on speakers of other languages, especially in the U.S. Even so, it annoys me to hear my language denigrated, insulted, and vilified. I do not think that English is inferior, and I do not think it is stultifying. What English speakers do is not a reflection on the language itself. A similar effect can be seen in anthropologists, most of whom argue passionately for religious diversity, as long as the religion isn't Christianity. I think we need to remember that no matter what terrible things English speakers have done to speakers of other languages, viz., boarding schools for Native American children, the language itself is just as valuable and just as beautiful as the languages I am, quite honestly, more interested in: Navajo, Lillooet, Karuk, Cherokee.

Plurals

In linguistics we often use the term "marked" to mean a structure or sound that is in some sense more difficult or less common across languages. The "unmarked" structure is the one that is default, typically in a cross-linguistic perspective. Thus you (to my knowledge) never find languages in which the present tense is derived from the past, or in which all obstruents are voiced and all sonorants are voiceless; the "default" tense is past, and the default voicing for obstruents is [-voice]. The term "marked" comes from literal morphological marking, i.e., past tense is literally marked in English by the suffix -ed, whereas the present tense is unmarked (in the first and second person singular). Likewise, singular is unmarked and plural is marked, in that languages add something to signify the plural, or don't change anything, but there aren't any languages (again, to my knowledge) that have an unmarked plural and then add an affix to derive the singular. This should correlate with frequency: unmarked forms and more common and marked forms are less common. Thus in doing a corpus based search for singulars and plurals, you should find more hits for singular forms of a word than plural forms, with a few exceptions for special cases like "pants" and "scissors". So I wondered the other day why I kept adding -s to things while I typing.

I noticed especially that I was doing it on the word "consonant" -- I kept typing "consonants" even when I meant the singular. So I decided to check out COCA (Corpus of Contemporary American English) to see if I was just weird. (Since this is a blog post and not a research paper, I haven't gone through the effort of determining the percentage of forms that are exactly what I'm looking for; thus the numbers for "consonant" below could include adjective usages as well as singular noun usages.)

• consonant -- 443
• consonants -- 323

So at least in this case I do seem to be an anomaly. It's not that "consonants" is used more often in English, like "scissors". Most likely I most commonly use the plural rather than the singular in my own (typed) usage. A word count check on my M.A. thesis confirms this: 75 counts of "consonants", but only 71 of "consonant". This may be because I rarely would talk about a specific consonant, but rather a specific phoneme, whereas I often have cause to talk about the natural class of consonants as a whole, as a subset of the phonemes of a language.

Just for fun, let's see some other COCA counts for singular and plural.

• computer -- 51,711
• computers -- 15,832
• woman -- 130,459
• women -- 211,930
• man -- 253,485
• men -- 157,413
• scissor -- 102
• scissors -- 1846

Those show some interesting patterns. "computer(s)" shows the expected pattern, with more than three times the hits for the singular than for the plural. However, we see an interesting difference with "women" -- more than 1.5 times more hits for the plural. My hunch is that this represents a similar pattern as my use of "consonants". People have little need to specify a singular person as a woman; they can just talk about a "person" named Mary. It's apparently when speaking about groups that gender becomes relevant. On the other hand, "men" shows the opposite pattern, with many more hits for the singular, just like "computers". My first thought would be that many of these are interjections: "Man, I'm tired", since much of COCA comes from spoken conversations. However, on looking at the actual results, it looks like very few are actually usages of this type. Another possibility is that many of these represent generic usages, like "the fall of man". But looking over the hits it appears this too is not very well represented, though there are some. I suppose we just have to chalk it up to the fact that in English masculine is the unmarked geneder, and possibly also that despite affirmative action women are still underrepresented in many sections of professional and academic life. "scissors", of course, patterns as expected: only a small minority of people use "scissor" in the singular. In fact, almost all of those hits turn out to be the adjective form, rather than the singular noun form.

Ejectives

I gave my students a brief introduction to IPA last week, and this coming week we're spending every class on a different aspect of phonetics, including IPA transcription. So I'm getting ready for a fun week of making silly sounds and writing with symbols that most people have never seen. One sound class that people seem to enjoy encountering for the first time is ejectives. They occur only in about 15% of languages, and most of those languages are minority languages that mainstream people never hear. In fact, they're exotic enough that Paul Frommer chose them as an element of the Na'vi language used in the movie Avatar.

Most inexpert descriptions of ejectives are mindbogglingly useless. If I tell you to pronounce /t'/ by pronouncing a more forceful "t" sound, I'll wager just about any amount of money or hat-eating that you're not going to come up with an ejective. The best way I've found to describe ejectives is to pronounce a consonant while holding your breath. After all, this is essentially what distinguishes ejective stops from regular stops: regular stops use the pulmonary airstream, while ejective stops use the glottalic airstream, without accessing air from the lungs. Ejective stops are the most common ejectives, but languages also have ejective fricatives such as /f'/ and /s'/, and Tlingit even has an ejective lateral fricative (it's a pretty tough one the first couple times). Though no language has ever come up with one of my favorite possible sounds: an ejective alevolar trill. Give it a try and see how many contacts you get before your tongue runs out of air.

Linguistics 101

I'll be embarking on my first official foray into linguistics teaching on Tuesday, with a summer section of LING 101 here at Rutgers. As opposed to 201, which is more focused on analysis, 101 is essentially an introduction to the kinds of things people talk about in linguistics; students are expected to come out of it knowing what assimilation is, but not necessarily how to write a rule describing it. I plan on focusing in part on why I think linguistics is interesting, and what kinds of things linguists do.

For me, one of the most fascinating aspects of language (with a lowercase ) is the variety we see in human languages. For many linguists this variety is a minor annoyance in their Universal Grammar exploits, something to be explained away and something that is ultimately superficial. My opinions about some contemporary formulations of UG aside, I think the variation among languages is something to revel in, and to explore. Most people are shocked to find out that there are almost 7000 languages spoken around the world (give or take a few), or that they've only been grouped into about 80 families, as opposed to the 2 or 3 I thought existed before I became a linguist. While some comparative linguists insist that there is evidence of a very small number of language families, people that actually work on those languages very rarely accept such a small number. While Greenberg posited only 3 families for the Americas, you'll be hard pressed to find an Americanist who accepts fewer than at least a couple dozen families, though work is being done to properly demonstrate genetic relationships between some of those families.

People are also sometimes intrigued by how different languages can be. As speakers of Indo-European languages we often assume that our way of doing things is the right way, or even the only way. But then you come across the Algonquian languages, where the ordering of person-marking affixes on the verb indicates not grammatical role, but simply the presence of that person in the action. The order of the affixes is fixed, and their person hierarchy combined with a large set of thematic verbal prefixes indicates which person is acting on whom. Many linguists still think every languages has nasals, because they haven't heard of Chemakum, Makah, Nitinaht, Lushootseed, or Twana (don't let the spelling fool you; pronounce all those as if you had a bad cold).

In short, I'm looking forward to six weeks of showing a group of students that not everything is like English, and that there are more languages in the world than are dreamt of in most naive conceptions of it.

Anglicizations

I did a double take when I saw that my last post was over a month ago, but I guess I have been that busy. At any rate, but got me thinking today was a Language Log post on the myriad (mis)pronunciations of the Eyjafjallajökull volcano (properly pronounced [ejafjatłajøkʰʏtł]). Obviously English speakers are going to have a hard time with the rounded front vowels and lateral affricates, so it seems worthwhile to have a general Anglicization.

Or does it? I guess I'm generally in favor of Anglicizations. I find it obnoxious when people pronounce Chile [tʃile] rather than the more standard [tʃɪli]. On the other hand, I routinely say Tlingit [łɪŋlɪt] and Nahuatl [naxʷatł] with the original laterals. I think the key issue is one of faithfulness to the original form versus creating an easily pronounceable form in the target language. In the case of Chile, only the vowels are changing, and the vowel changes are fairly minor: laxing of the /i/ and raising of the /e/ (the latter being part of the Great Vowel Shift in English). On the other hand, changing a consonant sound strikes me as more major. In cases where a spelling pronunciation of the foreign term is just too different, as in Eyjafjallajökull, I think it's best to simply pronounce the word as in the original language.

This might not be possible or useful for the average person, but I don't think it's too high a standard to hold newscasters to. It's there job to report on things, and I think it should also be their job to say things right. Sure, foreign words are a bit tongue-twisting at first, but a little practice and Eyjafjallajökull will be rolling off your tongue. And the more someone practices foreign sounds and foreign words, the easier it is not only to pronounce words in that language, but in other languages as well.

The conflict of faithfulness and ease of pronunciation of course shows up in much more unconscious speech as well. I'm in the midst of a project with James Crippen on exactly this type of conflict in Tlingit, a Dene language spoken in Alaska. Since Tlingit doesn't have the sounds /b/ or /p/, it's interesting to see how words are adapted when they are borrowed from languages that do have them.

Movement

I was watching an old episode of Seinfeld the other night, and a line of dialogue caught my attention: "That's the guy I told where the elevator was." It's not ungrammatical for me, but it's marked in some way that made my analytical skills perk up. According to current syntactic theory, this type of sentence is created by movement from one original, underlying position to a different surface position. You don't have to take this literally as movement; many syntacticians take "movement" as a relation rather than an actual move from one position to another. But the key is that a clause like "where the elevator was" is assumed to be in some way structurally the same as "the elevator was there". In some analyses of relative clauses, the same thing is true for nouns described by relative clauses, so that the DP "the guy" comes out of the CP "I told <the guy> where the elevator was". So the so-called D-structure (don't read any theoretical assumptions into that, I'm just using it as a convenient label) of "That's the guy I told where the elevator was" would be something like "That's I told the guy the elevator was where", changed by movement into "That's the guy I told <t> where the elevator was <t>".

But why did I find this remarkable? I'm fairly certain I use such structures. Before I analyzed the sentence, I thought it might involve some sort of movement where one constituent crosses over the trace of another (which, at least in Relativized Minimality, wouldn't necessarily involve a violation, but might be more marked in some way), but this isn't the case. The "the guy" movement occurs entirely in the matrix clause, and the "where" movement is entirely limited to the subordinate clause. So there's no crossing involved, just regular movement in two separate clauses. I tested out some similar, simpler examples, and found out that I find "I told him where the elevator was" completely unremarkable, but "That's the guy I told to clean up the mess" (very) slightly remarkable. (Side note: can we really have judgments about such minute differences? Maybe not everyone, but I've been dubbed "most sensitive speaker of English" at Rutgers, so I claim the right to make such judgments. I'll happily give different numerical scores to subtly different subjacency violations.) So what, I don't like relative clauses? It's not that, because "That's the guy I told" is also completely unremarkable. I think it's some very slight underlying preference for traces to be at the end of a clause. This would make questions unremarkable ("Who did you talk to <t>?"), as well as simple relative clauses ("That's the man I talked to <t>"). But more complex relative clauses with material after the trace make some small part of my language faculty slightly uncomfortable.

deep-end/depend

While perusing Not Always Right the other day, I came across a post entitled The Rule Deep-Ends On How Cute You Are. I was nonplussed; deep-ends? Certainly I'm familiar with the deep end of a pool, but as a verb? It took me the entire reading of the post to realize they were playing on the (in my opinion, mostly orthographic) similarity of "deep end" to "depend".

I think the main reason for my confusion is that I always, even in careful speech, pronounce "depend" as dǝ.ˈpʰɛnd, rather than di.ˈpʰɛnd. "deep end", on the other hand, is ˈdip.ˌɛnd. Thus, even if we treat both as single phonological words, "deep end" is different in stress, aspiration, and vowel quality. I can't tell if the assumed transparency on the part of the author is due solely to orthographic similarity, or if most people (or at least the author) have the unreduced vowel quality in the first syllable.

Lather, rinse, repeat

There was recently quite a bit of interesting discussion about a Language Log post on the semantics of "this page intentionally left blank". Levi Montgomery posted that it reminded him of "the instructions to 'Lather, rinse, and repeat,' apparently ad infinitum." I was struck by this comment, because historically and upon careful thought, I don't find anything recursive about this statement.

Levi apparenty interprets this injunction as being of the form (A --> B --> return to A). This would indeed lead to infinite hair washing, with the user lathering and rinsing until the eschaton. But I rather have always interpreted the instruction as having the form ((1: A --> B); repeat 1). Thus there's no recursive loop, merely a second execution of the two events A and B. Or to phrase it another way, for me there is no way of deriving wide scope of repeat so that it includes lather, rinse, and itself. For me it can only apply to lather and rinse. I'd be interested to find out if more people get Levi's interpretation or mine, and why.

Enamored

I was musing the other day about the PP (prepositional phrase) complement to "enamored". My intuition is that "enamored with" is more common in modern times, but that "enamored of" is the original and prescriptively "correct" usage. Checking with the OED more or less confirms this, though the original usage, unbeknownst to me, is "enamored (up)on". "Enamored of" was the next oldest usage, and "enamored with", though listed as a possibility, didn't have any examples.

Now to check modern frequency:
enamored upon: 833 ghits
enamored on: 6380 ghits
enamored of: 644,000 ghits
enamored with: 667,000 ghits

So it appears that my intuition was marginally correct, though with the inaccuracy of google results counting, there may be no significant difference between "of" and "with". More unexpected was the auto suggestion "enamored by", which gets 128,000 hits, less than the two recent usages, but far more than the original Middle English preposition. More surprising still is "enamored for", which gets a respectable 20,100 hits (though google enjoins me to correct it to "enamored of"). Many of these look to be merely a sequence, e.g., "names that mean enamored for girls", but there are some legitimate usages: "Armored and enamored for obama in DC". As a check I ran a couple other prepositions (under, from, beside) to see if in fact all are attested, but none of these three seemed to have any legitimate hits. So people using "enamored by" and "enamored for" seem to have that as the phrase.

What does it all mean? I don't know. But based on the auto suggestions from google, people are pretty unsure of which preposition to use, though "with" and "of" are by far the most frequent.

Vowels and consonants

I'm constantly struck by how many people seemingly refuse to believe in syllabic consonants. For instance, in my dialect (Standard American, or very close to it) there is absolutely no hint of a vowel in words like "word" or "bird". Yet many phonologists transcribe these words either as having a sequence of schwa plus /r/, or as the "r-colored schwa". I see no reason to posit any difference between the /r/ in the nucleus of "bird" and the /r/ in the onset of "rib". There surely must be a slight phonetic difference, but this is to be expected, because one is in onset position, while the other is in nucleus position. This is analogous to the slight difference between /u/ and /w/, or /i/ and /j/. One clue that it really is an /r/, and not a schwa plus /r/ sequence, or even an r-colored schwa: we get orthographic minimal pairs like "fur" and "fir" that are pronounced identically. This would be fine if they were clitics or unstressed syllables, where vowel reduction could neutralize both to schwa, but stressed "fur" and "fir" even in immaculate careful speech, are to my knowledge phonetically identical (I welcome any evidence to the contrary).

One reason people cling to the belief that syllables must have vowels is doubtless English orthography. Except for "rhythm", I can't off the top of my head think of any words that orthographically have a supposedly syllabic consonant (unless you want to count words like "icicle" that end with an orthographic vowel; feel free to post other examples in the comments if you find them). So in words like "butter", "bottle", and "button", where the second syllables contain a syllabic consonant, we still see a vowel in the written form (and presumably there was a vowel in the historical pronunciation).

Another reason is that we are taught, either explicitly or implicitly, that vowels are in some way the defining characteristic of a syllable. Many people are taught this in school, and become so dependent on orthography that some (native) speakers will even claim that the "th" sound in English (an interdental fricative) is a sequence of /t/ and /h/ (I swear I'm not making this up). However, orthography is always an imperfect clue to pronunciation, and English orthography is far from perfect, since its focus is on preserving the historical source of a word rather than transparently showing the pronunciation (NB: unlike many, I don't necessarily think that makes English orthography "worse" than a phonetically transparent orthography).

In English we have a limited number of syllabic consonants, viz., /n/, /m/, /l/, and /r/, i.e., sonorants. However, many other languages even use obstruents as syllabic. Berber and Bella Coola both utilize almost any consonant as a syllable nucleus; Bella Coola has entire vowelless sentences. What it comes down to is that there is no binary distinction between consonants and vowels; there is only the gradient sonority hierarchy, where sounds higher on the scale are more likely to be syllable nuclei, and sounds lower down are more likely to be syllable margins. For instance, within the five most common vowel quality distinctions, /i/, /e/, /a/, /o/, and /u/, /i/ and /u/ are classified as least sonorous, and are widely used (as /j/ and /w/) in syllabe margins, whereas /e/ and /o/ are more sonorous and rarely used as glides (as far as I know only in a few Papuan languages, and /o/ probably in Blackfoot), and /a/ is never a glide (unless perhaps /h/ is the consonantal version, an intriguing but questionable claim). For some languages, only the most sonorous sounds (vowels) are used as syllable nuclei, but other languages allow also the most sonorous consonants (sonorants) and others consonants lower yet (fricatives or even stops).

No update this week

I'm just posting this short notice this week because of new computer issues. Check back next weekend for a new post.

Xenolinguistics

Driving home from the LSA conference in Baltimore last weekend, I got to thinking about xenolinguistics, i.e., the study of alien languages. Obviously this isn't something we've encountered in reality, but it's often taken up in science fiction books and movies (most recently the movie Avatar). But of course in fiction the languages are always exactly like ours, and in many cases they even are ours, just slightly modified (e.g., Star Wars, The Dark Tower series). I think it would be endlessly fascinating to write a detailed treatise on what to actually do when confronted by an alien being, though I doubt any professional journal would be interested in the results. And the steps involved would not simply be "learn the language".

First you'd have to discover if they even had a language. This might not be as simple as it sounds, because they might communicate in a way that humans could not perceive (telepathy, chemical signals, EM emission with wavelengths below 380nm or above 760nm), or their communication might not immediately be recognizable as such. We make a lot of assumptions about universality, and many of these assumptions are not necessarily founded even for human languages, much less valid for Language in the abstract. Even in the simple case of a lifeform emitting some type of sound, you'd have to figure out whether or not it was communication, and then whether or not it was language. Noise emission could be unintentional, as humans radiate heat, or it could be an response to the environment or the lifeform's internal state. And if it does have a repeating pattern of some sort, the question would still remain as to whether or not it was limited to a finite series of fixed calls, as in monkeys and dogs, or whether it was an infinitely variable system of communication, as in human language.

Once language has been established, there's still the question of perceiving and reproducing the sounds in that language. An alien lifeform would doubtless have an acoustic production tract significantly different from humans, and thus it's not a given that humans would be even physically able to distinguish the sounds it produced, much less reproduce them. Certainly there would be ways around this. If the language is spoken in the 40k-80k Hz range, you could simply pitch shift a recording down to the human hearing range. More difficult might be a case in which the language was spoken in a very narrow range, say 300-350 Hz, with 10 or 15 tonal distinctions within that range. And there's no reason to think that the facts on phonology, syntax, or semantics would be anything like human language. Even if we admit all the tenets of Universal Grammar (which I find charitable), even a diehard UGer can't expect alien languages to be the same, since they'd have evolved a different Language Acquisition Device that could have different grammar rules in it, rules that might make no sense from a human language standpoint, or even from a human cognition standpoint.

I think it would be an interesting endeavor to write out a manual for the intrepid explorer encountering an alien lifeform for the first time, instructing him (or her) how to proceed in determing the presence of a language, and if it exists, documented and learning that language. But it doesn't seem like an endeavor that would appear very impressive to hiring committees or tenure review boards. Maybe I'll take it up in another fifty years or so, when I'm old and distinguished and I don't have to care what anyone thinks of me.

In Baltimore

This weekend I am in Baltimore for the Linguistic Society of America Conference, as well as the sister societies that meet concurrently. My primary reason for being here is to present a paper for the Society for the Study of Indigenous Languages of America session tomorrow morning. Being at a conference like this one brings to mind the primary things it means to be an academic, things that a lot of people seem not to understand. During this time of year, when fall classes have ended and spring classes haven't yet begun, I get a lot of comments from friends and family to the gist of "Well what are you going to DO with yourself?" Most of the general public seems to think that what academics do is limited to the classroom, and that when we're not taking or teaching classes we're relaxing, going for long walks on the beach, and spending time with friends and family.

I'd like to say nothing is further from the truth, but certainly there are things further from the truth. Obviously we like to try to do these things when our schedules are more flexible (i.e., we don't have many scheduled activities from day to day). However, people in academia do much, much, much more than just deal with coursework. In fact, coursework is often the easiest and least time-consuming part of being an academic. As a graduate student, I of course have my classes to worry about, but I also have outside engagements, some more social, such as department parties, potlucks, and coffee hours (which, while enjoyable, are nonetheless required), and some more professional, such as conference presentations, paper writing, and colloquia. Since I finished up with my fall classes (turning in my last assignment no earlier than December 21st, so not that long ago), my plate has still been filled with the following items: reading a lengthy paper for my phonology class that starts a week from Monday (yes, homework even over the break), finishing revisions for a paper to appear in conference proceedings, preparing my presentation for the SSILA conference this weekend (and of course attending the conference, etc.), submitting an abstract for a conference in March (which I may or may not have the money to actually attend), and writing an initial draft of a paper for another conference volume. But wait! There's more!

Those are just things that have a firm due date during the winter break. I also submitted a book review to a journal, and I'm continuing to work on two major projects: a journal article and the foundational research for a book. During the spring semester I will doubtless tackle new projects, some with deadlines, some just long-term research, mostly based on conference papers I haven't yet had time to expand into publishable material. Professors have all these same things to do, except instead of attending classes they're teaching them, which involves a lot more work. They also have to attend department meetings, review papers and grant applications for professional organizations, meet with students, serve as members or chairs of qualifying paper and dissertation defense committees, and seek to secure funding for their research. Don't get me wrong, I love academic life. I've worked 9-5 and (7-3) before, and it doesn't agree with me. Any salaried job tends to take up your entire life, so I'm glad mine is one that I'm passionate about and lets me be a little more flexible about when I do my work, even if that ends up being days, nights, and weekends. But it makes me sigh a little when people think we spend the summer tanning and the winter skiing. We don't. We spend the summer researching, and the winter researching. And the fall and spring researching.

Words for 'cat'

Frédéric Dichtel wrote me to ask about my research on words for 'cat', so I thought I would devote this week's blog entry to a summary of what I've done. I got interested specifically in words for 'cat' while researching neologisms in American languages (which I was researching in order to publicly support my private opinion that American languages more often create new words for new things rather than borrowing a word from another language). I noticed that, contra most other types of words, animal words were usually borrowed, often from English, but also other European languages. The word for 'cat' is a prime example of this, and displays probably more similarity cross-linguistically than any other word I looked at. My paperon the topic makes the claim that these similarities are due to a small set of widely diffused borrowings, rather than many separate instances of borrowing.

Similar words in different languages can be similar for four primary reasons: (i) the similarity is due to chance, (ii) the languages are genetically related, (iii) the form is borrowed, either from one language to the other or both from the same external source, or (iv) the words are similar due to some language universal. If languages are unrelated and in contact, the most likely scenario is usually (iii). Some examples are included below.

Language	Family	Word for 'cat'	Source
Mohawk	Iroquoian	takóós	Mithun (1999)
Munsee Delaware	Algonquian	póóšiiš	Swiggers (1985)
Mahican	Algonquian	póschees	Mithun (1999)
Blackfoot	Algonquian	póós	Frantz & Russell (1995)
Kootenai	isolate	pus	KCC (1999)
Chinook Jargon	Chinookan	pús(h)	James Crippen (p.c.)
Hanis	Coosan	puus	Grant (1997)
Klamath	isolate	p'oos	Barker (1963)
Umatilla Sahaptin	Sahaptian	p'uus	Thomas Morningowl (p.c.)
Walla Walla Sahaptin	Sahaptian	p'uus, pišpiš	Thomas Morningowl (p.c.)
Nez Perce	Sahaptian	pic	Aoki (1994)
Cayuse	isolate	picpic	Thomas Morningowl (p.c.)
Tlingit	Na-Dene	dóosh	James Crippen (p.c.)
Haida	isolate	dúus	Enrico (2004)
Coast Tsimshian	Tsimshianic	dúus	Dunn (1979)

Most likely these forms are from a combination of diffused borrowings from Dutch poes, the vocative form for 'cat' (i.e., how Dutch people address(ed) cats), English 'puss' and the English vocative 'psspss' used to call a cat. This would explain the prevalence of three types of forms: those that approximate poos, those that approximate pus (including almost all of the Salishan language, which I haven't included above), and those that approximate pispis. One piece of evidence that these are diffused borrowings rather than individual ones is that while Tlingit lacks labials, and thus has a form beginning with /d/ instead of /p/, Tsimshian and Haida have /p/, and thus must have borrowed the form from Tlingit rather than English or Dutch. James Crippen notes that the Tlingit form in turn is a borrowing from Chinook Jargon.

It's not clear what's so special about the word 'cat', but many northern languages have almost identical forms for this word, while other animal names are quite different. For example, the word for 'chicken' in the same languages displays remarkable variation, ranging from English borrowings to French borrowings to onomatopoetic terms to descriptive neologisms. Besides the northern languages, southeastern and southwestern languages also have very similar terms for 'cat', though they are from different borrowings (often old Spanish mozo or English 'kitty').