Mary, merry, and marry

For many speakers, including myself, "Mary", "merry", and "marry" are homonyms: meɹi. For other speakers, including my father, there is a threeway difference reflecting the underlying forms: me(j)ɹi, 'Mary', mɛɹi, 'merry', mæɹi, 'marry'. Generally, the distinctive forms belong to non-rhotic dialects, and the neutralized forms to rhotic dialects. This is because in rhotic dialects, intervocalic resonants tend to be ambisyllabic, i.e., they are attached both to the syllable that precedes them (as a coda) and the syllable that follows them (as an onset). An /r/ in coda position tends to neutralize many if not all vowel quality distinctions in the syllable it closes, and thus in rhotic dialects, where these syllables are closed by an /r/, we get all three front vowels neutralized to the [-hi][-lo][+ATR] vowel /e/. For non-rhotic speakers, /r/ can never be in coda position, and thus this neutralization does not occur.

Because of this, rhotic speakers tend not to be able to identify which form is which, even on hearing them produced by non-rhotic speakers (or rhotic speakers who happened to have picked up the distinction in careful speech). I occupy some sort of no man's land in between, since I understand the distinction, and can produce it, but I never use it in normal speech. I probably inherited this from my father, who, while a rhotic speaker, comes from family in New York, and probably heard many non-rhotic speakers (in addition to being a careful and conservative speaker himself). I recently encountered this difficulty on two fronts.

The first was in the TV show "Frasier". The character Niles, a rhotic but very careful speaker, played by David Hyde Pierce, also a rhotic speaker is discussing some former patients with commitment issues who overcame their disorder and were getting mɛɹid, which for a non-rhotic speaker would be "merried". This error seems a bit odd to me, since Pierce was born and raised in New York, was a camp counselour in New Hampshire, and went to school in Connecticut, so he surely was exposed to non-rhotic accents throughout his life. However, if he never acquired the distinction, it would be exceedingly difficult for him to recreate it. Though he didn't make the distinction, he knew that Niles likely would, and thus made a guess at one of the forms.

The second was my wife Amanda, discussing a coworker, with a New Jersey accent, who wished her a mæɹi Christmas. What the coworker actually said was almost certainly mɛɹi, but to a rhotic speaker like Amanda there is little, if any, perceivable difference between to two.

Na'vi

I feel like I have to comment at least a little on Na'vi, the language of the Na'vi people in the new movie Avatar. I haven't seen the movie yet, but I'm quite interested to do so soon, though I'm assured it does not revolutionize cinema as much of the press seems to claim. Paul Frommer was in charge of creating the language, and recently did a guest post on Language Log about some linguistic aspects of Na'vi. Since he's the creator, I see no reason to give a summary of the language (his description is well worth reading), but there are a couple things I can comment on.

One thing I like about Na'vi is that it uses ejectives. I don't like it for some scientific reason like "most languages have ejectives" (only about 15% do); I just like ejectives. They're fun to pronounce. I'm a bit nonplussed as to why Frommer chose to represent the glottalization with an "x", though: /p'/ is represented as px, /t'/ as tx, and /k'/ as kx. My guess would be that it makes the language look alien, which is as important a consideration as any when you're coming up with an alien language for Hollywood. Clearly Frommer has put a lot of thought into Na'vi; he even goes into restrictions on syllable structure. And these aren't just any random restrictions, but logical ones actually utilized in many natural languages. He remarks that only /f/, /s/, and /ts/ can appear as the first member of a consonant cluster. Now, I'm a bit dubious about the naturalness of this class in terms of actual occurrence, but at least theoretically it makes good sense; it's an exhaustive collection of the language's voiceless fricative phonemes (of course, /ts/ is not a fricative, but we can lump it under an ad hoc collection of "fricative phonemes" if we assume that affricates display edge effects, and since these sounds are the first members of clusters, the relevant edge of /ts/ would be the /s/ part). On the other hand, I don't think there really are any languages that do this. Some languages do allow only fricatives as the first members of complex clusters, but usually this is a class like /s/ and /hl/ (the lateral fricative; don't make me dig up my Unicode chart, I'm using Haida practical orthography), as in Haida. This is perhaps a more natural class because these are both coronal fricatives. Frommer's generalization is certainly theoretically warranted by some assumptions, because there seems to be something special about voiceless fricatives and clusters. But (as I argued in my M.A. thesis) it seems to be something special about coronal voiceless fricatives; I don't think we should expect to find, e.g., an extrasyllabic /f/ at the beginning of a complex cluster as we find extrasyllabic /s/ (claimed for English by Roca & Johnson, claimed for Blackfoot by me). On the other hand, for those who don't aspirate the /t/ in "fifteen", there's always the question of whether they syllabify it as fif.teen or fi.fteen.

Hopefully none of this seems like criticism of Frommer's language, because I certainly don't mean it as such. In a world populated by underdeveloped, Indo-European-influenced conlangs, it's nice to see someone as knowledgeable and dedicated as Frommer take the time to give us an artificial language that's interesting. I'd be interested in finding out a bit more about Frommer himself. All I can find is that he's in the business school as USC, and that he's referred to as a "linguist", but I haven't been able to dig up what linguistic research he's done, or what he looked at for his Ph.D. or M.A. (both from USC), if indeed either of those degrees were in linguistics. There are a lot more interesting things about Na'vi that you should read about in his Language Log guest post, but I've already run on for a while, so I'll end it here.

Could I get some binding indices, please?

This is actually a rather old news article, but I've only just now got around to blogging about it. The lede read "In an emotional interview, Whitney Houston opened up to talk-show host Oprah Winfrey about the pivotal role her mother played in getting the singer back on track." Wait, what? Whose mother? What singer? It's pretty unreasonable to expect that someone wouldn't be able to parse this sentence, but it still strikes me as very odd. First, there's the inevitable ambiguity in "her mother". This is resolved by pragmatics, since we assume that since Whitney Houston is the topic of the passage, "her mother" naturally refers to Houston's mother. However, syntactically this could just as easily be Winfrey's mother. (This is why I maintain we need a proximate/obviative distinction in English: "Whitney Houstonwa opened up to talk-show host Oprah Winfreyi about the pivotal role omotherwa played in getting oma singerwa back on track". When I mix Blackfoot and English I like to call it Blinglish.)

But the most serious "hey wait a minute" moment for me is the use of "the singer" toward the end of the sentence to refer to Whitney Houston. I would think that strictly speaking, this sentence should be ungrammatical by pretty much any version of the binding theory you adopt (classic GB, R&R, or Ken Safir's FTIP). This has to to with reference, which I think is captured nicely by Ken Safir's version of binding. Once we've established the context with an R-expression (referring expression, i.e., any noun phrase that's not a pronoun or an anaphor), we need to use the most dependent form for each successive instance of a coreferent NP. "Her mother" is fine, but it would be weird to say "Whitney Houston's mother"; that's why we have pronouncs. And "the singer" just seems really odd to me. I would expect "in getting her back on track". I assume they reverted back to an R-expression for Houston because it had been so long since the initial mention, but it's still very marked for me, perhaps even ungrammatical.

This is because you can't have an R-expression coindexed with a previous R-expression. If we say "Whitney Houston picked up the singer's clothes at the dry cleaner's", we want to ask "Wait, whose clothes did she pick up?" This violates whatever theory of binding you subscribe to, unequivocally. Luckily, actual language use is much more fluid, and clearly that was an acceptable sentence to someone, again, probably because of the distance between the two expressions, but I don't have to like it.

Copula contraction

I ran across an interesting instance of contraction on the Moviefone web site a while back, in the headline of a feature about some group of 80's stars or another: "Where They're Now". I found this interesting because you can't do this in English. Generally speaking, you can contract a copula onto the subject in English in an existential construction ("He is a good guitarist", "She is at the hospital") or when be is acting as an auxiliary verb ("They are going to the store"). This is reflected by a google search of "where they're now", which turns up millions of examples of constructions like "where they're now inside the city", but none of the Moviefone type.

This seems to be a function of wh-movement in this case. Note that the corresponding declarative "they are __ now" is perfectly happy to contract to "they're __ now". So why can't we do it after wh-movement? After all, we can say "they're happy" and "where they're happy". In all likelihood, this isn't a syntactic issue, but a phonological one, since contraction doesn't affect the syntactic status of the verb, only the phonological status. In a phrase like "He is a good guitarist", "is" is unstressed. Out of the blue, I have primary stress on "guiTARist", and secondary stress probably on "good". In "where they are now", on the other hand, "are" received some kind of secondary stress. I place primary stress on "now", but "where" and "are" both received secondary stress. It's for this reason that we (nominally) can't contract the copula onto the subject, because we can't get rid of secondary stress in that fashion. When there isn't stress on the copula, it can contract (or delete in ICE). I'm not sure if the Moviefone headline was written by a non-native speaker or just an overly efficient copy editor, but it's not well-formed in English, at least in my dialect.

Review of "The Life and Death of Texas German" by Hans C. Boas

When I was asked to review this book on my blog, I was unsure what I would find. Far from an expert on Texas German, I had in fact never heard of Texas German before received it in the mail this summer. However, as I've been slowly reading through it for the past few months, I've come to learn a great deal about Texas German and the rise and fall of this dialect. Overall, Boas' book is well-organized and extensively researched. His writing conveys a profound familiarity not only with the literature on Texas German, citing probably every major study undertaken of the dialecct, but also a keen interest in the process of language death, and the possibilities of language maintenance and revitalization. The only criticism I can offer is the rather clinical attitude he attempts to adopt in light of the death of Texas German, an attitude he clearly does not espouse, as evidenced by occasional glimpses of the author's true passion for the language and its continued survival. I found "The Life and Death of Texas German" to be an interesting work on three levels: (i) the analysis of Texas German as a language/dialect in its own right, (ii) the similarities of Texas German to many indigenous languages of North America in its current decline, and (iii) the origins and persistence of distinct American dialects of German, which is my own heritage language through my mother's bloodline.

The book is perhaps most obviously a useful resource for any researcher working on Texas German, or more generally on American dialects of German. More useful still is the Texas German Dialect Project, of which this publication is a product. The TGDP is a project undertaken by Boas with the help of a few research assistants to document Texas German before it becomes extinct. It has as one of its more important products the Texas German Dialect Archive. For his research, Boas developed several questionnaires ranging from translation tasks of words and sentences from English to questions about the informant's attitudes toward Texas German. (I should note here that Boas' use of the word "informant" is dated from my own Americanist perspective; generally we prefer to use the term "consultant".) Boas first gives sociohistorical context for the formation of the Texas German dialect, giving an overview of German immigration to Texas and the settlement patterns of the German settlers in central Texas, specifically around New Braunsfels, where Boas did his fieldwork for the TGDP. He then comments on new-dialect formation in Texas German, especially as regards Trudgill's (2004) model of new-dialect formation. Latter chapters give examples of specific developments in Texas German phonology and morphosyntax. Throughout, Boas argues that Texas German never underwent the final "focusing" stage of Trudgill's model, in which a dialect settles on a consistent pattern of phenomena (which is distinct from early stages which display significant interspeaker variability). In his final chapter, Boas comments on the impending death of Texas German and the possibility of language maintenance.

The parallels between the moribund Texas German dialect and the many languages of North America undergoing language death are striking. While the impact of the death of a dialect of a major language like German may not be as severe as the death of a unique language such as, e.g., Cayuse, the processes that languages undergo as they fall into disuse are fairly universal, as discussed in Fishman (1991). However, Boas does note that Texas German seems to retain its morphosyntactic features to a greater degree than is usual among dying languages. The reasons behind the decline of Texas German are all too common: status as a minority language, discrimination, lack of official legal status, disuse due to perceived economic and social advantages of the majority language. In the case of Texas German, the language enjoyed considerable prestige in its early days, when significant parts of Texas were entirely German speaking. This situation declined as roads better connected different areas of the country, causing an influx of English-only speakers into the New Braunfels area and an exodus of native Texas German speakers to bigger cities in order to find jobs. World War II played a large role in the branding of German as an "un-American" language, not only in the passing of English-only laws for schools and even some public spaces, but a decline in even private use by native speakers, who considered themselves Americans and did not want to engage in activities that were perceived as unpatriotic.

On a personal note, this book held my interest as a non-speaker of German, in that it is my heritage language yet I have inherited only three phrases from my German-speaking ancestors: was machst du, 'what are you doing?', nicht so laut, 'not so loud!', and gesundheit, 'bless you!' Both my maternal grandfather and my maternal grandmother were of German stock, the latter more recently than the former. However, none of my German-speaking ancestors passed their native tongue on to their children. Even my grandfather's great-grandfather Valentine Denzer, who was born in Germany, spoke English for most of his life, even keeping his personal diary in English during the Civil War. I think I too inherited this tendency: before I became a linguist, I was of the mind that if I had children in a foreign country, I would see no reason to teach them English, and that while I would continue speaking English to my family back home, I would use the local language during the rest of my life. Clearly this tendency stems from the desire for your children to have a better life than you had, and the belief that any deviation from the norm results in social difficulty and financial loss. It doesn't help any that this belief is at least somewhat accurate; while speaking another language is never a handicap, identifying first and foremost with a language or culture other than English can be a stumbling block in the United States. This same attitude has contributed not only to the decline of Texas German, but almost every indigenous language. In the case of indigenous languages, mandatory boarding schools, where children were beaten for speaking their native languages, certainly had an enormous impact as well, but in modern times, it is primarily the belief that identifying as English-speakers will help their children which keeps native speakers from passing on the language they grew up speaking.

"The Life and Death of Texas German" is a valuable resource for researchers in many areas of linguistics and anthropology. The Texas German Dialect Archive is likewise an incredibly valuable resource, especially since it may soon represent the last data available on Texas German. Boas offers a wealth of data on Texas German, not only on phonological and morphosyntactic phenomena that distinguish Texas German from Standard German, but also on speaker attitudes toward Texas German, including how often speakers used Texas German historically and in modern times. In many ways Texas German parallels the plight of indigenous languages of Americas, coming from a proud tradition of vigorous use, and falling into decline as English gained ground as the majority language associated with social status and economic advantage. Given the large percentage of readers who come from a Germanic background, Boas' book will no doubt also be of interest on a more personal level, with German as a heritage language which has been lost in many families. Boas' book is eminently readable and clearly written, presenting a valuable introduction to Texas German for the non-expert, as well as giving useful commentary on language death in general.

Apostrophe's

It's become quite common recently (unless this is the recency illusion striking again) for people to get confused and use apostrophes in plural forms, e.g., dog's for dogs. I'm not usually one to criticize non-standard usages, but this one has me puzzled. How do people get confused about this? 's is essentially never used in the plural, except for capitalized acronyms which haven't been lexicalized, and even then I think only MLA recommends using 's. So it's not a case of people being unsure when to use it for plurals and when not to; the rule is never use them.

So why is everyone so confused? The nature of this error makes it extremely difficult to research, since you have to hand pick the true instances, as opposed to the (still) more common genetive 's. I did find this gem through a google search: Hey guy's. I was wondaring do you love dog's or cat's? I like dog's!!! Please say dog's. Dog's rock! (Keep in mind that this was on what appears to be a forum for pre-adolescents and younger people in general.) In this case it seems the poster has internalized the rule as being that plural morphology in English is always 's. But as I said before, since this is never true, how to people get confused? My guess is simply interference from the genitive. I know from experience that people have trouble figuring out the difference between guy's and guys' (or guys's). Thus the confusion is not a grammatical one based on what plural marker to use, but a typographic one, in that people often see 's after a noun, and somehow they generalized it to plural marking.

I'd be interested in researching this further, but I'm at a loss for how to do a search for forms.

Structural ambiguity

A competitor in a Food Network show I watched recently was described as an "award-winning cake and sugar artist". Fairly straightforward, but my language faculty at first wanted to parse this is [[award-winning cake] and [sugar artist]] rather than [award-winning [cake and sugar] artist]. This is essentially the opposite of low attachment, so I'm not sure what was going on. Perhaps a desire for coordinated phrases to be coordinated as high as possible in the syntactic structure of the phrase.

Extended break

I'm not planning on doing any more updates until probably early next month, since I've started classes now, in addition to developing ESL materials for PronouncePro and working on some abstracts, papers, and a book review. Check back on October 3. If there are topics you're interested in hearing about this semester, feel free to post them in the comments section.

Two stories

I recently picked up the second season of the NBC police procedural show "Life" and have been watching it this weekend. In one of the episodes a crime takes place on an unnamed reservation in the desert near Los Angeles. As I watched, I tried to figure out what tribe it was supposed to be, mostly based on the language (after all, I am a linguist). What could it be? Western Pomo perhaps? I'm not familiar with many Uto-Aztecan languages, so I attempted to look the episode up online to see what language the actors were speaking during the few non-English lines of dialog. But suddenly I caught the word wašiču, 'white man' in Lakota (which, ironically, I learned not from my Lakota textbook or dictionary, but from my wife, who has a near-encyclopedic knowledge of the movie "Dances with Wolves"). This could be taken two ways. If we want to nitpick, we could get offended that the producers saw no need to use actual tribal members from the area or research the correct language for the tribe. Or if we want to be charitable we could be grateful that in a major network sitcom they actual decided to use Native actors speaking a Native language.

The second story is short, and comes from Bruce Rigsby via Phil Cash's Nez Perce mailing list. I've removed the name since I'm not sure if it would appropriate to reproduce here.

"Years ago several Old People on the Umatilla Reservation told me much the same account about ------, but it centred on the parable that it is easier for a camel to pass through the eye of a needle than a rich man to enter the Kingdom of Heaven. The old man meant to say quu'ys haama "rich man", but mistakenly said k'uuys haama "indecently exposed man"!"

Qeciyew'yew', Bruce, for showing us that Freudian slips don't just occur in Indo-European languages.

Why linguistics is useful when learning a foreign language

Betsy Lowe has been keeping me up to date on her endeavor to learn Hungarian, and the frustrations that go along with any attempt to learn a foreign language. She commented specifically on exceptions in vowel harmony, to which I replied:

"After a while those irregularites will start to resolve into patterns. For instance, my dad, who got his degree in linguistics, was at first puzzled by the Arabic definite article, which is usually al-, but the -l- changes to the first letter of the noun it attaches to in some cases. After a while he figured out that the -l- stays an -l- only when the first letter of the noun isn't a coronal consonant, and once he realized that he no longer had to memorize the cases; it was easy to figure them out."

As another example, in many languages velar stops become post-alveolar stops or alveolar fricatives before front vowels. Thus in Italian syllables originally beginning with /k/ now begin with tʃ, e.g., cibo, 'food' is pronounced tʃibo. In Italian the pattern is not difficult to remember even if you don't know any linguistics: c is k before a, o, and u, and tʃ before i and e. However, a little linguistics knowledge makes it even simpler: tʃ before front vowels, k elsewhere. Such knowledge is especially helpful in situations like the Arabic case, where without linguistic knowledge the learner merely has to memorize a long list of letters that take the article al- and another long list that take aC-, where C represents the first letter of the noun the article precedes.

The derivation of "Missoula"

As you know if you've read my little info box on the right side of this page, I currently live in Missoula, Montana. The derivation of the word name "Missoula" is somewhat opaque, but is generally agreed to derive from the Flathead (Selish) word nmesuletkʷ. However, not everyone agrees on the meaning of this word. I have heard perhaps most commonly from white people and historians "river of ambush/surprise", and from Native people "icy water", referring to glacial lake Missoula. Naturally I'm inclined to give more credence to the latter (though of course indigenous people are just as prone to folk etymologies as we Euro-Americans are). After doing some research it seems that my hunch was justified.

The easiest part of nmesuletkʷ to deconstruct (for a non-speaker of Salish) is the suffix -etkʷ, which means "liquid", often specifically in the sense of "water" in place names, cf. ntx̣ʷetkʷ, 'river'. The nmesul- part is a bit harder, but there are clues in several of the Salishan languages. The root sul seems to mean "cold" or "frozen": slsulčsti, 'his hands are freezing'; suł, 'froze'; cf. Spokane sul, 'cold'. The initial n- is presumably the locative marker present in many Salishan languages, including Spokane and Okanagan. The me- is the only part for which I was able to find an unequivocal answer, but may be a stem formative cognate with Okanagan -m-. So a rough translation would presumably be something like "place of the frozen water", quite likely a reference to glacial lake Missoula.

Vote for this blog!

I told myself I didn't care about contests like this, but clearly I was lying.



If you like reading this blog, go and vote for "Ryan's linguistics blog" at the above link.

Thanks.

Garden path sentences

I came across a headline recently that momentarily stumped me: "Judge accused of sex crimes impeached". This was because I automatically parsed this as "a certain judge has been accused of sex crimes", but then I came across "impeached" and had to fit it in somewhere". This happened because I initially interpreted "accused of sex crimes" as a verb phrase, whereas semantically it functions as a relative clause modifying "judge". These types of sentences, where we assign an initial interpretation and then have to revise it when we get to the end, are often referred to as "garden path" sentences, because we get led down a figurative garden path before getting to the actual meaning. A classic oft-repeated example is "The horse raced past the barn fell". Can't make sense of it? Try "The horse that was raced past the barn fell". (This doesn't work for everyone, because for some people racing a horse past a barn just doesn't feel right semantico-syntactically.) We initially think this is going to be just "The horse raced past the barn", with racing as the action the horse is performing, but when we get to the end we realize that actually the horse is falling, not racing, and "raced past the barn" is a relative clause modifying "the horse".

Onsets

Betsy Lowe emailed me an interesting observation last week, noting that in the song "The Devil Went Down to Georgia", the lyrics go "He drew his bow across the strings and it made a evil hiss" (not an evil hiss). She commented that she commonly hears this a-before-a-vowel pattern in native Southerners, and thought "that was incredibly unusual because Italian and the other Romance languages I've dabbled in, and now Hungarian, go to great lengths to PREVENT two vowel sounds from running into each other. E.g.: not 'e una altra cosa,' but 'ed un'altra cosa,' in Italian. In Hungarian, 'a' is 'the' or 'that', and if the following word begins with a vowel, then it becomes 'adz.'" This brings up an interesting point about syllable markedness, which is that syllables without onsets are marked, and thus many languages have epenthetic consonants or allomorphs to avoid a sequence of two vowels across a morpheme or word boundary. Besides English, Italian, and Hungarian, there are the Algic languages, which all have personal prefix allomorphs that are CV or CVC, depending on whether the stem begins with a consonant or a vowel. For instance, in Blackfoot the stem for 'dream' is paapáó'kaan, while 'my dream' is nipápao'kaani. The stem for 'boat' is aahkioohsa'tsis, but 'my boat' is nitááhkioohsa'tsisi. The 1st person prefix has allomorphs ni- and nit- to avoid onsetless syllables. Feel free to post examples from languages you're familiar with in the comments.

As a native of Atlanta, GA, I can't say I've noticed the prevalence of "a V-" pronunciations in Southern English, but as a native of Atlanta, GA, I didn't have much contact with natives of the deep South. Please comment if you have supporting or complicating evidence for Betsy's intuition. I'll close with a link to the song on YouTube, where you can hear that it really is əivl̩hɪs, rather than ejivl̩hɪs or əʔivl̩hɪs. The part in question is around 1:20.

More syntactic constituents

I've talked about syntactic constituents before (here), and recently another difficult regarding them came to my attention via the Famous Dave's web site. On their site you are prompted to "enter either a zip code or select a state". If this were proper VPE (Verb Phrase Ellipsis), I would be expected to (i) enter a zip code or (ii) enter select a state. Since (ii) is ungrammatical, clearly something has gone wrong here. My guess is that this phrasing resulted from a blend of (a) "enter either a zip code or a state" and (b) "either enter a zip code or select a state". Note that with (a) we get (ai) enter a zip code or (aii) enter a state, and with (b) we get (bi) enter a zip code or (bii) select a state. However, as is the request would be parsed as [enter [either [a zip code] or [select a state]]].

Different to

I've noticed that the narrator of the Discovery show "Mythbusters" uses quite a few idiosyncratic phrases. By idiosyncratic I really just mean different from what I perceive as the standard and common way of saying things; I could be wrong. One of these I noticed the other day was "different to" something. I think in pretty much all cases I would say that one thing is "different from" another, though I might could use "different than" as well. Google gives the following results:

• different from: 128M


• different than: 47.5M


• different to: 10.9M

So apparently I'm not crazy in my ranking of different from >> different than >> different to. From those google searches I also noticed that different to is apparently common in UK English, but rare in US English. I can't think of a good way to check frequencies within a given dialect on google, but I feel like different to may be more Southern. This fits with the Anglophilia of Southern English, as well as the fact that I think the Mythbusters narrator is Southern, based on some non-standard syntax and pronunciations he uses once in a while.

Syntactic constituents

I recently saw a KFC ad that gave me pause: Mix it in your bucket. Why is "mix" highlighted? My guess is that their ad campaign consists of a number of similar slogans, with the initial verb highlighted. Okay, so why did this strike me as odd? Because "it in your bucket" isn't a constituent. A syntactic constituent is, narrowly, a group of words which is entirely and exhaustively dominated by a single node, i.e., there is some syntactic node which dominates all of and only that group of words. More practically the domination doesn't have to be exhaustive: we could certainly say that "mix it" is a constituent, even though the IP node also dominates "in your bucket". But "it in your bucket"? Not even close. The most general parsing of the sentences would be [mix it][in your bucket], and most narrowly would be [[mix [it]][in [your [bucket]]]]. There's no way to derive a constituent "it in your bucket".

Linguistic pet peeves

Plenty of people have linguistic pet peeves, phonetic or syntactic variants, usually non-standarad, that they cringe upon hearing. I'll admit to my fair share of them: I have a slight adverse reaction when I hear people say "I wish I was..." or "This is him". But I'll tell you what really bugs me: people who insist that their linguistic pet peeves should be enforced upon everyone. When I hear someone say "I wish I was..." I think to myself "I wish I were..." But I don't say anything, and I don't think to myself, "Man, that person is an idiot." Language is what people say, not what's in the Strunk & White. This isn't to say that all prescriptive grammar is hogwash. Some of the prescriptive suggestions really do lead to clearer, better writing. But so much of it is just linguistic peevery. What's wrong with, "I wish I was"? Is it ambiguous? Is there something that the subjunctive adds to the meaning of the sentence? No. It's merely agreement. In fact, the subjunctive is so useless that it shows up in languages very late in their development, and it's often one of the first features to be lost. That's not to say I don't love and use the subjunctive; I just recognize that it's a personal thing, not a linguistic law.

Pronunciation of final -e

My wife and I were last night discussing the varying pronunciations of Nietzsche: nitʃi versus nitʃə. As far as I'm aware, the latter is the preferred pronunciation, based on the fact that every philosopher I know pronounces it this way (and they're the ones with most cause to use it), and that Nietzsche was German, in which a final e such as this one should, according to my scant knowledge of German, be pronounced as a schwa. So why do people say nitʃi? That's what I said until I started taking philosophy classes. My hypothesis is that it stems from final -e in the numerous Greek words and names we've borrowed into English. To give some examples from linguistic and English terminology, "apocope", "syncope", "synecdoche", or from Greek mythology, "Persephone", "Ariadne", etc. etc. When people look at a word they don't say "Hmm, what language is that from?" They have vague unconscious knowledge of how they've heard other similar words pronounced, and in this case that results in a translation of the German -e in Nietzsche to the final -e in Greek which is pronounced i. Another reason we should have linguistics classes in school.

Another argument for the necessity of language education

I was perusing Amazon.com the other day when I happened upon this gem in a review for a Sepultura album: "Surely destined to become yet another Sepultura classic, A-LEX (Russian for no law) will catapult the Brazilian masters right back into their deserved spotlight." The translation of "a-lex" seems acceptable, but from Russian? Wikipedia claims that this is from Latin ab-, 'away from' + lex, 'law'. I think more likely it's a mixture of Greek a, 'without' + Latin lex, 'law'. I would argue that this is another reason why we need linguistics education for all. I think it's reasonable to expect anyone going through even our current education system to have some basic understanding of where certain common prefixes come from, but a little bit of training in linguistics would surely benefit students even more.

Some people would no doubt question the usefulness of such training: who cares if we know the derivation of Sepultura's album title? Well I say fie on them. Ultimately you can argue the same thing for any bit of knowledge.

Person hierarchy in English?

Arnold Zwicky wrote today about agreement with disjunctive subjects, i.e., clauses in which the subject is separated by an operator such as 'or' or 'nor'. He gives examples such as Neither Barbara nor I {am, is, are} able to... and If you or I {am, are} here.... This phenomenon is one that most if not all people struggle with, and I believe it provides solid evidence against the model of the brain as a computer which simply spits out answers based on some strictly written grammar code.

There are several competing motivations here. In the former example, one of these for many people is to say things that are prescriptively correct. This would be Neither...is, because prescriptively, 'neither' is the "subject" of the sentence. Of course, most of us also want the verb to agree with the most immediate NP, in this case, 'I'. Yet Neither Barbara nor I am... definitely sounds strange, because 'am' doesn't encompass Barbara. I think I would probably choose 'are', on the basis that this particular sentence is semantically equivalent to something like We are unable to..., i.e., the subject is in some underlying way a 1st person plural. On the other hand, it is difficult to apply this kind of view to the second example. I wouldn't want to say 'am' for the same reason as above, but I also wouldn't want to say 'are' because of the conflict with the 1st person pronoun 'I', and the fact that because of the disjunct the subject is clearly singular.

I think that in some ways this can be explained by some resort to a person hierarchy in English. We typically want 1st person pronouns to come second, as in Jim and I went to the store as opposed to ?I and Jim went to the store. But this hierarchy creates problems because of the above examples, where putting 'I' first would probably result in a more natural sentence (from a verb agreement standpoint).

more low attachment

I realize there haven't been any updates in a while, and the only thing I have to offer by way of an excuse is the ~90 pages of M.A. thesis I've written in the past 8 weeks. Now that I have the bulk of the chapter drafts finished, I should be able to get back to weekly updates.

Today I want to talk some more about low attachment. Low attachment is something that's been discussed on Language Log from time to time, and refers to the tendency of listeners to process a given phrase as attached to the lowest node possible (roughly, that a given coordination is at the same level as whatever immediately preceded it). Today's sample came from a warning on my school email account, regarding several phishing attempts that had been perpetrated recently. It listed several email addresses and included the warning, "If you receive any messages from these sources, do not open them and delete them."

Because of the tendency toward low attachment (surely someone needs to formulate a constraint based on this), my initial reaction was to parse this as [do not [open them and delete them]], i.e., ~(A U B). So perhaps I could open the messages, or delete them, but not both. Of course the actual warning was ~A U B, i.e., that I should not do the former, and that I should do the latter. A rough parsing of this would be [[do not open them] and [delete them]]. But this requires attaching the second VP to a higher node than the first reading, which is why I was initially confused by the message. It's extremely rare that some kind of parsing problem like this ever results in momentary confusion, but it does occasionally make for amusing examples.

The epiglottis qua severe throat infection

I was amused to see mentioned in an article the other day "epiglottis (severe throat infection)". The author was referring (or attempting to refer, depending on your opinion of how reference functions) to epiglottitis, inflammation of the epiglottis, which I imagine is indeed a fairly severe throat infection. So why the misspelling? (Note: I'll admit I had to look up whether there are one or two s's in "misspelling". While the answer turned out to be two, "mispelling" is actually more common, at least in terms of ghits.) My guess is copy editing and/or spell checking. Epiglottitis just looks too look, and like it has way too many t's. My guess is that -titi- sequence threw someone off. Who knows if it it was a person or a computer program, but either way it resulted in a rather humorous typo. Guess I'll have to wait until my next bout with strep before I can practice my Haida. (Obscurity rating: 10/10)

mail vs. email

I was struck the other day by the difference in both the noun and verb forms for "mail" vs. "email". First, let's talk about the nouns. Mail is strictly a mass noun: we say I got some mail, not *I got a mail. Email, on the other hand, is different in two ways: (1) it's generally a count noun, but (2) not strictly so. While it's more common to say I got an email (1,390,000 ghits), we can also say I got email or I got some email (I'm not putting a ghit # for these because it would take forever to tease out, in the former, non-native speakers leaving out the article, and, in the latter, referential as opposed to partitive some). Email is a thing that you can send: I sent him an email. Mail seems less so. To me, I sent him mail implies that perhaps a roommate that moved out continued receiving mail at my house, and so I forwarded it to him. Less likely to me is the reading that I sent him something by mail.

Now on to the verbs. For mail the primary object is the thing being sent, not the sender: I mailed him a letter, I mailed a letter, ??I mailed him (meaning I sent something to him). If I say I mailed him it sounds more like I put him in a box and sent him something than that I sent him a package. Email is the reverse: the primary object is the receiver of the email, not the message itself. I emailed him an article is fine, as is plain I emailed him, whereas ?I emailed an article is grammatically fine but pragmatically odd: who did you send it to? (Note: I really like using whom, not because I like to be pretentious, but because it's part of my native grammar, but "to whom did you send it" sounds just too stilted for me to utter it in public.)

Mail and email also seem to differ with regards to telicity. At least for me, "I'm emailing him right now" indicates that you are in the actual process of emailing, i.e., the process of emailing someone consists not only of hitting send, but the writing of the message which leads up to that point. It is one of Vendler's accomplishments, whereby an action consists of some activity leading up to some culmination point. "I'm mailing a letter to him right now", on the other hand, seems to only admit the possibility that you are about to mail the letter. Mail seems to be one of Vendler's achievements: a singular event with no activity leading up to it. You can't be said to be literally in the process of mailing a letter. Either you haven't mailed it yet or you have.

If you...then it

President Obama said in an interview with CNN (I think at the time this was President-elect Obama), "If you think about the journey that this country has made, then it can't help but stir your heart." However much I may agree with his sentiment, I can't help but be put off by the phrasing.

I think the problem for me is the use of "then". With if-then sentences, I seem to want the subjects to be coreferential. Without "then" they don't need to be at all. "If you think...it can't help..." sounds fine to me. But "If you...then it" strikes me as off somehow. As far as I know there's no prescriptivist rule regarding anything like this. In fact, I'm sure prescriptivists would always want us to include the "then", citing some nonsense about ambiguity or form. It's not "then" I have a problem with either, because "If you eat now, then you won't be hungry later" is fine, because the subjects are coreferential. But for some reason my language faculty doesn't like non-coreferential if-then sentences with an explicity subordinator. It's a mystery.

A new area of responsibility

I was struck by a CNN headline a few days ago that mentioned Obama's presidency and "a new area of responsibility". Obviously they meant "era", but "area" was in both the headline and the text of the article. It was corrected within a few hours, but left me wondering if the writer actually thought or meant "area" of if it was a typo.

I seriously doubt the writer meant "area". A "new era of responsibility" is certainly a phrase that's been going around recently, so "area" must be a typo. I tried a bunch of different misspellings in Word, but didn't get any that list "area" as a suggestion. One thing I typed, that I was unable to recreate, resulted in the sequence "a rea" or something similar being automatically changed to "area", so that's one possibility, but it rests on leaving the "n" off of "an", misspelling "era", and furthmore leaving out "new" and just talking about "an era of responsibility". That seems like too many steps to me.

Another possibility is that it was a phonological mishap because of the pronunciation similarities between "era" and "area". This happens to me all the time. My fingers often type what I hear in my head as opposed to what I'm actually thinking, so that while I would never mess up there/their/they're in a paper, I often do so in quick IM typing. I once answered "know" to a polar question.

This blog believed read, I say

If you're anything like me, you may have had a bit of trouble interpreting the title of this post to reflect my belief that people read this blog. I modelled this odd phrasing after a headline that caught my eye the other day, about the pilot of a plane who bailed out in Florida after falsely stating that his plane was crashing: "Mystery pilot believed found, authorities say". I think the problem here is some kind of collapsed double passive construction.

The original statement is something like "It is believed that the mystery pilot is found", in turn reflecting some statement on the part of authorities like "We believe we have found the mystery pilot." The first sentence, with two passives (one in the main clause and one in a subordinate clause), isn't that difficult to comprehend. But when you don't have any expletive subjects, articles, or auxiliaries, it's a bit difficult to parse.

I think another part of the difficulty is that we are loath to interpret "found" as an adjective. For me "mystery pilot believed dead" isn't nearly as bad a sentence as "mystery pilot believed found". It's those two passives crammed together without any auxiliaries that does it in for me.

Thank you for selecting

Here in Missoula the movie theater choices are fairly limited: two Carmike theaters, and the Wilma theater, in the old Wilma hotel in downtown Missoula. I've always been struck by a line on the Carmike cinemas intro screen: "Thank you for selecting Carmike Cinemas". Let me tell you why I find this a little annoying.

To me selection has to do with being presented equivalent options and choosing one of them based on certain criteria. A dictator is not selected, and if I'm starving in the woods and can't find any food, I'm not selecting berries if I stumble across a blackberry patch. Likewise, the movie choices in Missoula, though there are multiple venues, don't really allow for selection. The Wilma only shows indie films, and the Carmikes only show mainstream releases. Furthermore, the two Carmike theaters usually show different movies so that they aren't competing for the limited business in town. So if there's a given movie you want to see, chances are good that there is only one place to see it. I don't consider that selection.

Obviously not all markets are as small as Missoula. However, many markets are smaller than Missoula, and Carmike, as far as I know, only builds theaters in rural areas or suburbs of smaller cities. It seems that their business model is built on the premise of limited competition in out-of-the-way places. So I wonder if people ever "select" Carmike Cinemas. "Thank you for choosing" would be much more felicitous for me, and I'm not sure why. In terms of denotation, "choose" and "select" are essentially the same: you are presented with options and you pick one of them. Yet for some reason "Thank you for choosing to eat berries" wouldn't be as infelicitous in my above described survival scenario. Perhaps it has to do with "Thank you for choosing..." as a more set phrase in our society, whereas "Thank you for selecting" is essentially purely compositional for me, e.g., "...choosing..." for me is like "blue ribbon" (a coherent concept in our society), whereas "...selecting..." for me is like "green ribbon"; it doesn't really mean anything other than the sum of its parts.

If I were X and Y

Today's topic comes from an episode of Friends. One of the (male) characters is discussing a hypothetical situation and says "If I were a man..." and then trails off, the audience laughs, and he continues, "Did I just say 'If I were a man'?"

The intended sentence, of course, would have been, "If I were a man and X", where X is some hypothetical situation. This is an example of a counterfactual, where the protasis (the if clause) contains a hypothetical situation which is counter to fact. The sitcom character takes the counterfactual in his dialogue to be "If I were a man", i.e., "If X" where X = "I am a man". In fact, the counter to fact clause is "If I were a man and X", where X is the hypothetical situation in question. Since this is a counterfactual, we know that the protasis must be false, but the protasis is not "If X", but rather "If X and Y". While the opposite of X is ~X (where ~ is the negative quantifier), the opposite of (X and Y) is ~X or ~Y, not necessarily ~X and ~Y. Thus in the dialogue, the speaker can of course still be a man and say "If I were a man and Y". All this means is that one of the conditions be false, in this case presumably Y.

The mismatch between haplogroup and language

Language is an important part of culture, identity, and everyday life, so it makes sense that we want to collapse language and culture/ethnicity. However, it is almost never the case that linguistic and genetic boundaries line up exactly, certainly not in modern times, and relatively rarely even in ancient times (as far as can be determined). Don Ringe recently did a fascinating guest post on Language Log on the linguistic diversity of aboriginal Europe. He notes specifically the quick spread of Indo-European languages, with the result that "while most Europeans’ linguistic ancestors were speakers of PIE, many or even most of their biological ancestors at the same time depth were speakers of non-IE languages already residing in Europe."

Nowhere is this more evident than the modern United States. With the exception of the 1% or so who speak an indigenous American language, virtually all of the 300 million American citizens speak English (even if it is only as a second language). Even immigrants who learn English poorly or not at all tend to have children who are at least bilingual, if not monolingual in English. This same kind of dominant language spread most likely occurred in ancient Europe, and indeed all over the world. Yet often anthropologists and occasionally linguists like to attempt to tie genetic groups to certain languages.

One famous example of this is Greenberg's three language groups in the Americas. While most linguists who study indigenous American languages allow around 70-80 language families (at the low end; many people insist on many more), Greenberg claimed on the basis of his language comparisons that there are no more than three "stocks" in the Americas: Eskimo, Na-Dene, and Amerind. While the first two are recognized families, "Amerind" lumps together the rest of the 80 or so language families spoken from Canada to Chile. Most linguists, especially historical linguists, object to Greenberg's style of classification because it relies on shallow, wide surveys of languages rather than narrow, in-depth analysis. Much of Greenberg's evidence for relatedness comes solely from the frequency of /n/ in first-person markers. Much more of his data lists cognates between words which have different numbers of morphemes, or are clearly borrowings.

Greenberg's theory is often "supported" by those who point to the three genetic groups in the Americas, which rougly correspond to Greenberg's three linguistic families. The error here is in thinking that because two peoples belong to the same genetic group, they speak the same language (or even related languages). Now, it may be the case that Chapakuran languages are related to Algic languages, but if they are, the relation is so distant that we will never find evidence of a link unless we develop time travel. Glottochronology (which relies on assumptions and rates of change rejected by all but the most die-hard language lumpers) predicts that after 15,000 years, two related languages will share about 6-7% of their vocabulary -- approximately the same as chance resemblance. Mark Rosenfelder discusses this at length in his article How likely are chance resemblances between languages?.

So that's not to say that it's incorrect to equate haplogroups and linguistic stocks, just that it's not falsifiable. It may well be that there was a single Proto-World language from which all languages are descended. However, because of the great time depth at issue, this is a matter for faith, not for science.

I wish I would have...

One of the common "errors" that I've been noticing recently is the use of "I wish I would have" (and similar constructions) for "I wish I had" (and similar constructions). Of course in reality there's nothing wrong with this construction; it's simply not the proscriptive one. My guess is that it's not even the less common construction.

A quick check on google gives the following results:

• "I wish I would have" - 5,080,000
  
• "I wish I had" - 19,900,000
  

Okay, so the nonstandard construction is indeed less common, but only by about 4:1. And five million hits is enough to question why people use this construction. Saying "that's how they learned it" isn't explanatory, because this just shifts the question back a generation, to why their parents said it that way, ad infinitum. My bet would be on the desire for a different construction marking the protasis of a conditional clause (the protasis is "if" clause of a conditional construction; the apodosis is the "then" clause). While language does tend to eschew redundancy, there are so many cases of redundancy in language(s) that this is clearly a constraint which is readily violated. The phrasing "I wish I had gone to the store" is perhaps a bit puzzling if we take out the conditionality. Why "I had gone to the store"? Why not "I have gone to the store"? After all, it's a wish about the present situation. This may be why some people prefer "I wish I would have gone to the store", because it's transparent. "I would have gone to the store" makes more sense out of the subordinate context, and simultaneously shows the conditional/irrealis nature of the wish in a way "I wish I had gone to the store" does not.