Val Systems

Why does Labov have such weird transcriptions?

noreply@blogger.com (JoFrhwld) — Sat, 21 Jul 2018 18:50:00 +0000

It is the experience of many linguists, when they crack open something written by Bill Labov, or his students, that they suddenly realize they are using a transcriptional system that they are probably not used to. Sure, the symbols <i> and <e> look familiar, but when you realize they correspond to the IPA symbols <ɪ> and <ɛ>, you might start to feel uneasy. The worst offender, though is <oh>, which is the vowel that appears (in some dialects) in thought, lost, law. What is that? Why is there an <h>in there? Why isn't he just using IPA??

What are these symbols?

The Labovian transcriptions has the following symbols, (presented with the corresponding Wells' Lexical Set).

These symbols were largely adopted from Trager & Bloch, and as they outline in the ANAE, apparently had some widespread usage in American dialectology & phonology in the early 20th century. Trager & Bloch developed this system largely to make a phonological argument about English: Contrary to commonly held belief, English only has 6 vowel phonemes! The details of that analysis isn't too important, but what does matter is that each "subsystem" forms a natural class of sorts.

Why not use IPA?

Well, I think there's a good reason. To begin with the simplest answer, in sociolinguistics, we often need to make reference to categories more abstract than phonemes. The Labovian notation is actually a diaphonemic one, and the use of IPA symbols, as they're conventionally understood, wouldn't be quite accurate. For example, what's the IPA symbol for the vowel that appears in cat? The obvious answer would probably be /æ/. But let's look how that vowel is distributed in the Atlas of North American English.

The proper IPA transcription for the word that appears in cat seems to vary between [a̠~a~æ~æ̝~ɛ̞]. Ok, well, that's in the narrow transcription, you say. In the broad transcription, it would be /æ/. But still, I say it's not so clear cut. Looking at the relationship between vowels in the Inland North, I don't really think it would be correct to give the vowel in cat the broad transcription /æ/.

We could give the Inland North its own phonemic analysis, and give cat something like /ɛː/ in broad transcription, but we would start to seriously befuddle ourselves doing cross dialectal comparisons. Or even intra-dialectal comparisons, if you're doing a study like Monica Nesbitt, tracking the reversal of the Northern Cities Shift.

Why not use Lexical Sets?

So, there is a system of denoting "word classes" in a way that could avoid diaphonemic confusion: Wells' Lexical Sets. Using Wells' Lexical Sets, we could say that across North America, the word cat has the Trap vowel. This label, Trap, denotes a class of words that have a common linguistic history such that they often have a common phonemic incidence in contemporary dialects. So, regardless of what the phonemic analysis of the Inland North really is, cat has the same vowel as hat as sad as mask, etc.

If we could use a lexical set system this way, why don't we? The answer from the Atlas of North American English is that lexical sets are insufficiently structural. For example, what is the phonological relationship between Fleece, Kit, Face and Dress? Well, as far as the transcription system goes, these are just lists of word classes, with no particular phonological relationship implied except that they're not the same vowel.

In the Labov/Trager/Bloch system, these vowels are represented as /iy/, /i/, /ey/, and /e/. The transcriptional system is now encoding a phonological analysis. Because /i/ and /e/ are single characters, they are short vowels. Because /i/ and /iy/ share a character, they share a height and backness specification.

Essentially the system constructs a systems of analogous contrasts. As /iy/ is to /i/, so /ey/ is to /e/. And as /iy/ is to /ey/, so /i/ is to /e/. In a sense it's a very stripped down feature system encoded in the transcription.

Ok by WHY is <oh> that way!?

Well, ok. This part was inherited from Trager & Bloch, and was then kind of screwed up by the push to eliminate special characters from the system. It makes enough sense to transcribe the Lot class as <o> . First off, in British English and some North American dialects this vowel is rounded. Secondly, most words in the class are spelled orthographically with <o>.

Next, Trager/Bloch/Labov decided to represent a long and ingliding vowel as ending in <h>. That's because they were trying to define all non-short vowels as ending in a glide, and with /y/ and /w/ already claimed, /h/ remained.

In the phonological analysis of Lot and Thought, Thought minimally contrasts with Lot in terms of this /h/ glide, so it gets <oh>.

The real problem is <ow>! In the phonological analysis, /ow/ does not share height and backness with /o/. The low back vowel with a /w/ glide is /aw/. If the Labovian system hew more closely to the Trager & Bloch system, it would represent the Goat vowel as /ʌw/ (they represented it as /əw/). But, one principle Labov strove for was "minimum deviation from Roman typography" (Labov et al 2006, a.k.a. ANAE).

Which system should be used?

To be honest, I think all three transcriptional systems ought to be engaged, so long as they are properly indicated. I think the Labov/Trager/Bloch system forces viewing the vowel system of various dialects with a certain systematicity that is clarifying. An IPA transcription is necessary for the phonetic detail, which is not clearly communicated with simple formant values. Finally, when speaking to an audience, I think you should always use the Wells' lexical sets.

If you did a study analyzing variation in (Labovian notation now) /i/ and /e/ varied, I think the following two ways of saying them are confusing:

I found an effect for [aɪ] but not for [iː].
I found an effect for [ɪ] but not for [ɛ].

The first option could confuse people into thinking you were analyzing /ay/ and /iy/. The second option is going to be perceptually difficult, if only because English doesn't allow stressed lax vowels. People won't be confused about which two vowels you're talking about if you say:

I found an effect for Kit but not for Dress.

Brussel(s) Sprouts

noreply@blogger.com (JoFrhwld) — Mon, 18 Dec 2017 16:49:00 +0000

YouGov has come out with a perplexing, Christmas themed¹, survey. "Sorry Britain," they say, "but you've been saying 'Brussels sprouts' wrong." Apparently 77% of respondents think the little green cabbage looking vegetables are called "Brussel sprouts." But as they point out, these vegetables get their name from the capital of Belgium, Brussels, and are original called "Brussels sprouts."

Sorry Britain, but you've been saying "Brussels sprouts" wrong: 77% of Brits think the correct plural term for Christmas's most famous vegetable is "Brussel sprouts". Only 18% got it right https://t.co/hNsN7VV6x8 pic.twitter.com/verEN5w3YA
— YouGov (@YouGov) December 18, 2017

Ok, there's a lot going on here. First of all, what does it mean for 77% of the speakers of a language to "get it wrong"? If not only most, but the vast majority of people in the UK think these vegetables are "Brussel Sprouts", then what other basis is there for determining whether or not that is correct? Of course, we could turn to etymology, but that kind of approach to determining "correct" language is going to produce really weird results. Imagine this takepiece, if you will:

When you got dressed this morning, did you put a shirt on the top half of your body and skirt on the bottom half? If you agree with 100% of English speakers that these are two different words for pieces of clothing, you're not alone, but you're all WRONG! "Skirt" and "shirt" both come from the Old Norse skyrta! Why did people start ruining the language and mispronouncing skyrta to mean different garments? The Oxford English Dictionary says this an "unexplained difference of sense," probably because when you say "shirt" and "skirt" differently you stop making sense!

Pretty nonsensical, right? What's more, a shift from "Brussels sprouts" to "Brussel sprouts" is a very natural kind re-cutting of word boundaries that has happened many times in English. Oxford dictionaries has a nifty blog post about some of these cases involving "an." For example, when you set out to make your next big meal (maybe even a big Christmas dinner), you might tie on your kitchen apron. But the word used to be "napron". Getting from "napron" to "apron" wasn't just a case of people forgetting to put the /n/ at the beginning. Unlike with writing, when you speak, there's no spaces between your words. If you said:

I put on a napron.

It would come out of your mouth something like

I put on [əneiprən].

And then, a listener would have to figure out where the boundaries between the words are. But there's a two different options that could both work.

If you go with the first option, you decide that what what has been said is "a napron". But if you go with the second option, you decide what has been said is "an apron". At some point over the course of English, enough people went with the second option and the word became "apron." Are we all doing it wrong?

But the weirdest thing to me about YouGov's tweet was that they say we've been saying Brussel(s) sprouts wrong. First off, they only surveyed people about how they spell the vegetable, not how they say it. And second, if one person was saying "Brussels sprouts" and another was saying "Brussel sprouts," how would you even tell the difference?

I actually recorded myself saying "Brussel sprouts" twice and "Brussels sprouts" twice. Do you see a glaring difference in the offending [s]?

Ok, one of the "Brussels sprouts" [s] is about 60 milliseconds longer than the rest, but the point is that people don't speak letters out loud. We regulate our breathing and move our tongues around in a way that bears no resemblance to an "s", or "s s" as it's written or typed on a page. Asking people how they spell these words doesn't really tell you about how they pronounce them, and even if someone intended to say "Brussels sprouts," it would probably be indistinct from "Brussel sprouts" anyway.

Anyway, in closing, roast your Brussels sprouts, don't boil them. They're better that way.

1 That is, Christmas themed in the UK, where Brussel(s) Sprouts are part of the traditional Christmas meal.^↩

A Scottish Narrative of Personal Experience

noreply@blogger.com (JoFrhwld) — Sat, 11 Feb 2017 18:36:00 +0000

In a web extra from Samantha Bee's show, we get an extended interview with the farmer Michael Forbes telling a story about how he chased Donald Trump Jr. away from his Aberdeenshire farm. It was actually an excellent example of a narrative of personal experience, very similar to the paradigmatic type I was taught about in my sociolinguistics classes.

It's been a long time since I've tried any kind of narrative analysis, so apologies for any errors, but I'm going to give it a shot here cause I think it's such a good example, and indicative of how rich even simple narratives like this can be.

Check out the video, it's worth it.

Here's a transcript of the narrative, with the contributions of the interviewer simplified a bit:

a. I chased his son once from here.
b. (Which one?) The one who's got the greasy hair.
c. (Which one?) Young Donald is it?
d. I was having a cup of tea.
e. And they knock knock knock on the door.
f. And eh, my mother answered it.
g. And she says "We've got visitors".
h. And she shut the door
i. So they tapped a bit louder, you know?
j. And, mother answered it again.
k. She says "I told you, we've got visitors"
l. Shut the door again.
m. Rattled louder.
n. Well, I answered it the next time.
o. I says "Get the fuck out of here"
p. I says "If you come back here again"
q. I says "I'll have you charged with harrassment"
r. "Don't get angry mate, don't get a--"
s. I says "Fucking angry"
t. I says "I'll show you fucking angry"
u. And they go out the gate.
v. (You loved it) Yeah I did, Yeah

This satisfies the Labov & Waletzky definition of a narrative in that the sequence of clauses is the same as the temporal sequence of events, except for line (a). Lines (a-c) would be called the "Abstract", outlining the most reportable event of the narrative. After we have the abstract, we have the orientation in (d) I was having a cup of tea. Then it is just complicating action after complicating action until the resolution in (u) And they go out the gate.

One interesting thing about this narrative is it is devoid of any evaluation. All of the turns are devoted to what events happened, and what was said, but there is no turn contributing what his state of mind was, or any other kind of evaluation of the situation. You could imagine the addition of a "It was rude to keep knocking" turn, for example, but that is absent. I was told once that avoiding evaluation makes for a better narrative for the listener, and is more typical of working class narratives. The interviewer, Amy Hoggart, tries to extract an iota of evaluation out of him (You loved it), but he doesn't repeat the evaluation, he just agrees with it (Yeah I did, yeah).

The repetition of the knocking and answering events is really interesting as well. They serve the obvious narrative function of heightening the tension, to good effect as we can see on the face of Amy Hoggart at turn (n) when Michael finally answers the door. I would hazard a guess that if Donald Trump Jr. knocked on the door twice, or four times, this story would (and should) still be told with three knocks. It's also interesting that in the first two knocking-answering events, DTJr and his entourage don't actually say anything. They knock, they are rebuffed, and the door is closed. There's no other exchange of words.

There's also a complex piece of cultural information being conveyed here, regarding when it is appropriate to call on someone. Michael's mother indirectly tells DTJr to go away twice by saying We've got visitors. No visitors were mentioned in the orientation of the narrative, so I'm assuming that there were no actual visitors in the house. The fact that you should not call on someone when they have visitors is presented as so obvious that it is a de facto instruction to leave within the narrative, and in the telling of the narrative it goes unexplained.

Finally, his verbs of quotation are awesome. He exclusively uses say in the historical present, and it seems like he is only reporting on speech which was said. D'Arcy (2012) has argued that the new verbs of quotation, like "be like" have been buoyed up by an increasing tendency to include quoted thought and mimesis in narratives. Mimesis isn't absent from Michael's telling of the story, but it is absent from the text. He physically acts out each knocking and each door shutting event, and he acts out chasing them out the front gate.

All in all, a pretty good narrative.

Tina Fey Nailed The Philly Accent

noreply@blogger.com (JoFrhwld) — Mon, 10 Oct 2016 11:13:00 +0000

Saturday night, Tina Fey and Jimmy Fallon appeared on Weekend Update as "White Women from Suburban Philadelphia." Apparently suburban white women are the "the swing vote within the swing vote" which was the premise for having them on the show. It seemed like it was also at least partially to make fun of Jimmy Fallon's softball treatment of Trump.

I'm blogging about the sketch in appreciation of Tina Fey's performance of the Philadelphia Accent. First off, you should watch the sketch:

Over at AV Club, they criticized both Fey and Fallon for their accents:

Despite how the two pronounced “hoagies,” the performances (and accents) were all over the place, with Fey not bothering to do hers once she launched a Fey-esque attack on Mike Pence’s anti-gay, anti-woman agenda.

The sketch might not have been funny, and Fallon clearly didn't know what he was doing, but Tina Fey was on point, in my professional opinion. It's not surprising Fey should be able to perform the accent since she's originally from Upper Darby, which is right next to Clifton Heights in Delaware County. That corner of DelCo actually has an interesting place in the study of the Philadelphia Dialect. R. Whitney Tucker wrote one of the earlier descriptions of the dialect in 1942 while he was at Pennsylvania Military College, now called Widener University. At the time, he said

I think that the real heart and centre of this [Philadelpia] dialect was originally, and still is, a few miles to the south, in the eastern part of Delaware County.

So how did Fey (and Fallon) deal with the accent? I previously blogged about Chris Matthew's native performance of the dialect by running through a list of the dialect features, but this time I ran the sketch through the FAVE-suite! So up ahead is a detailed phonetic analysis of Fey's performance.

Overall Vowels

First up, here's Fey's over-all vowel system. I look at plots like these all the time, so this is inherently meaningful to me, but I'll try to break it down a bit.

There's a few top-line things I see here. First off, she's moved her /ahr/ vowel (as in start and far) way up to be right next to her /ɔ/ (you can hear it when she says "in charge"). That is exactly correct, but her /owr/ could be higher, since pore, poor, pour are all merged in Philly, and is usually the highest backest vowel in the system.

She's also got a nice separation of /ay/ (as in ride) from /ay0/ (as in write) in a Canadian Raising pattern (you can hear it when she says mice). This is one of the more established features of the Philly accent now. There's also an /ey/ (as in face) and /eyF/ (as in gay) separation. This is less commonly discussed, but before consonants, eight is pretty similar to eat in Philly, while word finally (like gay) it's much lower, almost southern sounding. But, when I look at Fey's /ey/, /eyF/ difference in detail, it's not especially consistent.

One thing that looks like she's overly doing it is /uw/ fronting (as in scooter). She has it as far front as her /Tuw/ (as in do). Philadelphia is known for its /uw/ fronting, but usually it is much fronter following coronals than it is elsewhere, so I would expect to see /uw/ further back here.

She's also got some kind of /ɛ/ /æ/ merger that I can't account for. That's not a Philly thing that I know of. One last thing that caught my ear is some kind of consistent /ʌ/ backing (as in Trump). That isn't usually described as a feature or ongoing change of the dialect, but it sounds authentic to my ears.

Vowels in Detail

But let's get into more of the guts of the system. In the over-all vowel plot, there's a really good split between the tense and lax short-a (/æh/ and /æ/ respectively). But the Philadelphia split-æ system is complicated, but partially overlapping with related systems. So how'd she do in detail?

Based on limited data, I'd say she more or less nailed it. Her /æ/ before /s/ in jackass is tenser than the one before /k/ (although, a friend on facebook said they didn't think it sounded authentic). Her /æ/ in grandpa is a bit lax, but I think that's forgivable given the preceding consonant-liquid cluster. Most impressively, she's got a properly lax [æ] in her two repetitions of Indiana, and in banging. I would expect those to be the most likely to be messed up by someone trying to do an impersonation.

Incidentally, I think it could be the fact that she correctly had lax [æ] in Indiana that the AV Club thought she stopped doing the accent, since most American accents would have tensing there.

Honestly, this distribution of data makes me think that she's just got the split system natively. I haven't tried to compare this performance to something less affected, so I can't tell if she just always does this.

I also looked at her vowel dynamics a bit.

A few things jump out. First, I think she did a really good job on the phonetic quality of /aw/ (as in clown). I've found that it's basically a falling diphthong, and the most advanced tokens have a later F1 maximum, which she pretty much nails here. She also has a clear Canadian Raising pattern between /ay/ and /ay0/, with maybe some monophthongization of the pre-voiced tokens that's producing a strange trajectory. It's also possible that she's doing the /ey/, /eyF/ split in the formant dynamics, but I wouldn't put too much stock in that.

Here's what her /ɛ/ /æ/ and /æh/ vowel dynamics are like:

Clearly very good dynamic differences between /æ/ and /æh/. I included /ɛ/ here (labelled "e" in the plot) just to get another look at that weird /æ/~/ɛ/ merger. It seems to be there more or less in the dynamics as well. I really don't know what's happening there.

Consonants

I also did some quick and dirty coding of Fey's consonants. One of the things I noticed is that Fey was doing quite a bit of (dh) stopping (a dental stop in words like the and this). When I tallied it up, it looked like it was actually about 20% of her tokens, which is a bit low for Philadelphia.

	stop	deleted	fricative
count	4	4	13
proportion	0.19	0.19	0.62

She also had only one /str/ sequence, but she backed it to a very clear [ʃtr] in street.

I haven't tried to touch her /l/ vocalization or darkening at all, cause I think I have a tough time with that perceptually, and I don't have a fancy script for analyzing it like I do for vowels.

Fallon

Meanwhile there's Jimmy Fallon. I'm not going to go into detail with him, cause as he said in the sketch, his accent was all messed up. Here's his overall vowels, and his dynamics.

Yeah, so it looks like he's just vacating his back vowel space, and also monopthongizing a lot of things? Basically, this isn't anything.

Conclusion

In conclusion, Tina Fey nailed it, and Fallon didn't know if he was coming or going.

Open Question: "I'm a big fan of yours"

noreply@blogger.com (JoFrhwld) — Fri, 02 Sep 2016 15:24:00 +0000

Linguists, I've got a question. Is there any distinction to be made between (1) and (2)?

I'm a big fan of Rhianna.
I'm a big fan of Rhianna's.

I find both acceptable, but think I think I would prefer (1)? For pronouns, though, I think they must be possessive pronouns.

*I'm a big fan of you.
I'm a big fan of yours.
*I'm a big fan of her.
I'm a big fan of hers.

But I think it's strictly animacy based, since for inanimates I think the possessive form is ruled out, or is at least worse.

I'm a big fan of coffee.
*I'm a big fan of coffee's.
I'm a big fan of Star Trek.
*I'm a big fan of Star Trek's.
I'm a big fan of it.
*I'm a big fan of its.

I don't think the acceptability of (1) is related to a sort of corporate entity reading of "Rhianna." Both of the following seem fine to me:

I'm a friend of Joel.
I'm a friend of Joel's.

Thoughts?

The Sound of Silence

noreply@blogger.com (JoFrhwld) — Mon, 15 Feb 2016 19:56:00 +0000

One counterintuitive thing about doing linguistic analysis is how much time we spend analyzing structures and elements that are silent. That's probably pretty confusing for people when they first learn about it. For example, compare these three sentences:

I want a cookie.
I ate the cookie.
I ate cookies all day.

In your grammar lessons in school, you probably learned that the words a and the were called "articles," but linguists usually call them "determiners." Most people describing the sentences in 1, 2 and 3 would say something like:

"Sentences 1 and 2 have determiners in them. Sentence 1 had a and sentence 2 has the. Sentence 3 doesn't have a determiner."

A linguist, on the other hand, when describing these three sentences would be likely to say:

"All three sentences have determiners. Sentence 1 has a, sentence 2 has the, and sentence 3 has a silent determiner."

Silent determiners is just scratching the surface of all of the possible silent words linguists have postulated. A really common kind of reaction to all the silent words is "Bullshit!" Actually, that was my reaction when I took my first syntax course, but eventually, I was convinced.¹ For a lot of the silent elements linguists have proposed, there's usually some good reason or evidence for doing so, but we do have to be careful not to over-hypothesize silent elements to make the data work. This is the really interesting tension of abstractness.

Here's an example I'm dealing with in my own work, involving how Philadelphians have traditionally pronounced their short-a in words like mat and man. Usually, Philadelphians have a "tense" or "nasal" sounding short-a when the sound following is an /m/ or an /n/. For example, man is tense, but mat is lax. But this only happens if the /m/ or /n/ is in the same syllable. So the word ham comes out tense, but the word hammer comes out lax, because the /m/ is in the next syllable, if you sound it out (ha-mer). Plan is tense, but planet is lax (pla-net).

One weird exception to this pattern is the word exam. Exam usually comes out lax, even though the /m/ is in the same syllable when you sound it out (ihg-zam). One way of analyzing this is just to say "Ok, exam is just an exception to the rule." But, I'm going to make a different argument, which is that every time a Philadelphian says exam, they're actually saying an abbreviated form of examination, and in examination, the /m/ is not in the same syllable (ihg-za-mih-ney-shun). So the word exam is lax, because it's really examination.

One objection to this argument is that examination looks like it's exam+ination, so how do we know that what people are saying is examination and not just exam without -ination added to it. Well, let's look at some other words that end in -ination. We have imagination. If we abbreviate imagination, like "I've totes got a wild imag," we get out ih-majh which means the same thing as imagination, and it has stress on the second syllable. This looks like exam (ihg-zam), which means the same thing as examination, and has stress on the second syllable. What happens if we never add -ination to imagination? We get out image (ih-mijh), which doesn't mean the same thing as imagination, and has stress on the first syllable. That doesn't look like exam at all. This makes it seem more likely that exam is an abbreviation like "I've totes got a wild imag", and is not just a bare word, like image.

But there is no word ehg-zum, so does that mean that the root exam can only ever appear attached to -ination? That's not so weird if you look at other words that end in -ination, like destination. Dest never shows up as its own word. But it seems to show up in related words like destined or destiny.

There's also a little bit of historical evidence that exam was originally an abbreviation. If you look at some of the earlier examples of exam in Google books, authors used to end it in a period, like you would for any abbreviation. For example, J.M. Barrie (more famous for writing Peter Pan) in his 1889 book An Edinburgh Eleven wrote:

I knew a Snell man who was sent back from the Oxford entrance exam., and he always held himself that the Biblical questions had done it.

That period followed immediately by a comma makes it pretty clear that in this sentence, exam is an abbreviation. And if you look at Google ngram patterns, exam seems to be increasing in frequency relative to examination.

So exam looks like it was historically an abbreviation of examination, and the fact that Philadelphians have traditionally pronounced it with a lax short-a suggests that it still is an abbreviation for us, unconsciously.

This is just one example of how language is not a "what-you-see-is-what-you-get" game. There is a lot of silent structure to language that we're not consciously aware of. It takes a mixture of clever reasoning and empirical data to work it out.

But I'm guessing there's still a few people saying "Bullshit!" out there.

^{1. I actually once wrote a post on my undergraduate blog about how I thought PRO was absurd, and Norvin Richards left some really nice and patient comments on it. I'm kind of embarrassed about that now.↩}

#NWAV44

noreply@blogger.com (JoFrhwld) — Thu, 29 Oct 2015 15:43:00 +0000

I'm just recently back from the New Ways of Analyzing Variation (NWAV) conference, the top variationist sociolinguistics conference in North America. This was NWAV44, hosted by the University of Toronto, and as usual it was a lot of fun, and really exhausting.

When you go to a conference regularly enough, you start making conference friends, people who you only know and really only ever see in the context of the conference. Seeing my conference friends again is always one of the things I look forward to when going to NWAV, and it's what makes it all the more disappointing when I can't go for some reason or another. Of course, nowadays, if you can't make it to NWAV in person, you can always follow along at home on the Twitter hashtag. Really, it's almost like a parallel conference going on on Twitter. I go to a lot of conferences that get tweeted about, but I feel like NWAV tends to have a much higher rate of twitter traffic, and this was especially true of NWAV44. The Manchester LEL twitter account speculated that this might be the most tweeted about conference of all time.

Manchester at #nwav44 - possibly the most-tweeted linguistics conference of all time. https://t.co/3qOCRb0Vro pic.twitter.com/2kbYSTHlCZ
— Manchester LEL (@UoMLEL) October 27, 2015

It definitely spawned the most parody accounts (@GreatHallBird, @nwavAVghost). But, I thought I'd take this up and pick over the Twitter traffic on the #NWAV44 stream. I pulled down all the tweets I could with the twitteR package, excluding retweets.

In the 4 days of the conference, there were 3,196 tweets on #NWAV44. Here's the distribution of tweets, binned into 10 minute intervals, color coded by what was happening at the conference at the time (according to the official schedule).

The highest number of tweets in any 10 minute period was 53, for a rate of 5.3 tweets a minute, during the second paper session on Saturday.

There wasn't much tweeting during the poster session, which is too bad. First of all, it means a bit less exposure for people in the poster sessions. Second of all, posters are an intrinsically visual medium, and conventional wisdom is that tweets with pictures attached get more attention. I can't tell whether a tweet has a picture attached in this data, but I can tell if it has a link. I broke tweets down into two categories: tweets with an "https" link and does not contain the words "slide" or "talk", and all others. Winds up looking like this:

might have image	n	average retweets	average favorites
yes	378	1.29	3.31
no	2,818	0.82	1.89

Maybe something to think about next time you're live tweeting a conference.

Another thing I was interested in was what the average tweeting rate was during any given 20 minute talk + 10 minute question period. So, I took each talk period, and counted up how many tweets were sent each minute. Here's what it looks like.

During a talk, it looks like there's a pretty steady average of 2 to 3 tweets per minute persisting into the Q&A period, dropping off precipitously during the lat 5 minutes of Q&A as the speakers switched over. It's really pretty striking that during the meat of any given talk period, the most common number of tweets in a minute is something like 1 or 2, not 0.

One last thing I looked at was tweets about the birds. In the Great Hall in Hart House, there were at least two birds flying around the rafters and eating the crumbs after coffee sessions. It was a major topic of twitter conversation, and spawned the parody account @GreatHallBird. The volume of tweeting about the birds reached its peak on the third day of the conference, when about 8% of all tweets contained the string [bB]ird.

There was initially some competing ways of referring to the bird. Some people originally decided to call it "Ferdinand", but midday on the 24th, the @GreatHallBird account started tweeting, and eventually that standard became the most popular. Here is what people were calling the birds out of the options of just "[bB]ird", "[fF]erdinand" and "[gG]reat\s?[hH]all\s?[bB]ird"

Unfortunately, I don't have access to historical twitter data to check how #NWAV44 compared to other large conferences, like maybe #ICPHS2015.

UPDATE: I just realized that I didn't filter out tweets by the @GreatHallBird account itself when estimating the rate of tweets about the birds. When I exclude its tweets, as well as any line initial @-mentions, nothing really changes that much qualitatively, but the third day of the conference tops out at about 7% tweets about the bird, instead of 8%.

Gender in the Wasteland

noreply@blogger.com (JoFrhwld) — Wed, 05 Aug 2015 13:58:00 +0000

One thing I do with the bit of leisure time I have is play video games. I feel like I need to be a bit self defensive about it, given my age and place in life, but according to surveys the average age of gamers is somewhere between 30 and 37, so I fit right in there (although, the age distribution probably has a strong leftwards skew, so it'd be nice to know the median age). Many mainstream games have socially problematic themes, like violence being the only available recourse to progression in the game. A really cool video series on that point is the Grand Theft Auto Pacifist. Another common problem games have is their portrayal of gender (the topic of this post), and on that topic, you should obviously go watch Tropes vs Women in Video Games.

I think these socially problematic themes are a bigger deal than when they show up in other media. Video games are unique in requiring an alignment of the audience's motivations with the main character. For example, the experience of watching Breaking Bad and observing Walter White's moral descent would be very different from playing a Breaking Bad game and controlling Walter White. What I've learned playing video games is that while it may be projected on your TV screen, the game is really in your head.

I've recently been playing Fallout Shelter, a mobile game set in the post nuclear apocalypse Fallout universe from Bethesda Studios. Fallout 4 is maybe the most highly anticipated game coming out in time for Christmas this year, and Fallout Shelter is fun diversion for fans to play in the meantime. Your role is the overseer of a Vault-Tec vault where dwellers escape the radioactive wasteland. You build out and populate your vault, and assign dwellers tasks like food, water and energy production.

It's kinda like an ant colony.

Our story begins when I assigned a dweller to the Science Station I'd built so she could produce RadAways (a treatment for the omnipresent radiation). I wanted to equip her with the Professor Outfit, which would boost her Intelligence stat, speeding up the production of RadAways. I scrolled through the inventory a few times, and couldn't find the Professor Outfit. The only possibility I considered was that I'd forgotten to unequip it from the dweller who was wearing it before. But as I messed around, it became clear that you're unable to equip female dwellers with the Professor Outfit... Yeah, I know.

Here are two dwellers, Alexander and Judy, in the Science Station.

When I select Alexander and scroll through the outfits inventory, the Professor Outfit is right there between the Nightwear and Radiation Suit. When I select Judy, it's just absent. It's not even greyed out, just invisible.

This is a problem. There is a very strong cultural expectation that "Professors are Men", and this expectation gets visited upon my female friends and colleagues in really unfortunate ways. The most common and obvious day-to-day experience I've been told about is people assuming that "Prof Smith" is a man in e-mails. Or it's assumed that they're administrative support staff instead of faculty at meetings. Sometimes people assume female profs are students showing up for the first day of class, instead of being the instructor. One friend of mine said they got student feedback on a course saying that they were a great teacher and would doubtless be successful "in whatever career she pursues."

These are all obvious examples of women not being taken seriously in their professional roles, and the "Professors are Men" expectation has some obvious impacts on their careers. Women are grossly underrepresented at the highest faculty levels, and get paid about 90% of what men at equivalent levels do. So what's it matter that in Fallout Shelter, Alexander can wear the Professor Outfit and Judy isn't even presented with the option?

An all too common scene.

Well, first, it's just re-enforcing the "Professors are Men" expectation. Who is this message reaching? For one, me, and probably many of my students, and we're the ones who do a lot of the damage with the "Professors are Men" assumption. But it's also probably reaching a larger portion of women than even other games with poor gender portrayals do. It's a mobile game, and mobile games are disproportionately popular among women. Also, it has an undeniably The Sims-like element to it, a game which was also disproportionately popular among women. So, it's a fairly negative message about women, in all likelihood being disproportionately directed at women.

This is also a game with a bit of cultural reach. There is some speculation that it's out earning Candy Crush Saga, and it topped the App Store charts for a while. Bethesda is also a really big and popular game studio, and Fallout Shelter bears the "Editor's Choice" badge.

Preventing female dwellers from equipping the Professor Outfit is also the only gender based equipping restriction I've come across in the game. All of the other outfits can be worn by male and female dwellers. Both Alexander and Judy can wear the Combat Armor, but it's drawn a bit differently for their different bodies.

Sometimes, with things like gender representation, there is an in-game explanation for why things are the way they are, and we need to have a conversation about why gender representation in fictional worlds is an important issue for our real world. But I don't think that's what's going on here. The fact that the outfits are drawn differently for male and female dwellers, and the fact that the Professor Outfit is just unceremoniously absent without any kind of in-game explanation suggests to me that they just didn't bother drawing a Professor Outfit for female dwellers. That is, we're observing here a real world instantiation of the "Professors are Men" assumption rather than some kind of intentional fictional representation of that assumption.

But just because it isn't intentional doesn't mean it isn't sexist, and doesn't mean it shouldn't be changed. Being unintentionally overlooked is exactly the problem my female colleagues are facing. I've sent Bethesda an email briefly outlining this, and asking them to rectify it in an update. Don't know if I'll hear back from them, or if I'm better off yelling at no one in the Wasteland.

Edit: The pregnancy mechanics are also pretty messed up too.

Once pregnant, otherwise-competent women will become hysterical and run away from danger while flailing their arms. pic.twitter.com/NgKmae4Kqe
— Feminist Frequency (@femfreq) July 2, 2015

Video games are a broad medium though (this is more self defensiveness), and can be utilized for all sorts of purposes. For example, the Iñupiat Cook Inlet Tribal Council had the game Never Alone made, in which they embedded parts of their oral traditions for their younger generation. It's a really pretty and fun game. There's also the episodic game Life is Strange, which is essentially a game about friendship, family, the near constant threat of vitimization women face, and the anxieties we all face in trying to make the right choices in life.

A Silly "Name" Generator

noreply@blogger.com (JoFrhwld) — Thu, 04 Dec 2014 18:15:00 +0000

tl;dr: It looks like names aren't well modeled as a Markov process, but you can install my R package that does model names as Markov processes and mess around with it.

I don't know how I wound up writing a "name" generator yesterday, but I did. And now it's an R package (just on github for the moment), so you can play around with it too (https://jofrhwld.github.io/nameGenerator/).

I was messing around with some other research questions when I decided to see what would happen if I tried to model given first names as a Markov process. Here's a picture of a Markov chain that contains just the letters of my own name: Joe.

First, you start out in a start state. Then, you move, with some probability, either to the o, j, or e character. Then, you move, with some probability, to one of the other states (one of the other letters or end), or stay in the same state. In this figure, I've highlighted the path that my own name actually takes, but there are actually an infinite number of possible paths through these states, including "names" such as "Eej", "Jeoeojo", "Jojoe", etc.

A Markov chain for all possible names would look a lot like this figure, but would have one state for every letter. Now, I keep saying that you move from one state to the next with "some probability," but with what probability? If you have a large collection of names, you can estimate these probabilities from the data. You just calculate for each letter what the probability is of any following letter. So for the letter "j", you count how many times a name went from "j" to any other letter. For boys names in 2013, that looks like this table.

from	to	count
j	a	176452
j	o	84485
j	e	26118
j	u	25616
j	i	1920
j	h	425
j	r	121
j	c	118
j	d	98
j	end	55
...	...	...

As it turns out, a whole bunch of name data is available in the babynames R package put together by Hadley Wickham. So, I wrote a few functions where it estimates the transition probabilities from the data for a given year (from 1880 to 2013) for a given sex, and then generates random "names", or just returns the most probable path through the states. How often does this return a for real name? Sometimes, but not usually. For example, the most probable path through character states for boys born in 1970 is D[anericha] with that"anericha" bit repeating for infinity. For boys born in 1940, it's just an infinite sequence of Llllllll...

So, that introduces a problem where the end state is just not a very likely state to follow any given letter, so when generating random names from the Markov chain, they come out really really long. I introduced an additional process that probabilistically kills the chain as it gets longer based on the probability distribution of name lengths in the data, but that's just one more hack that goes to show that names aren't well modeled as a Markov chain.

Here's a little sample of random "names" generated by the transition probabilities for girls in 2013:

Elicia
Annis
Ttlila
Halenava
Amysso
Menel
Seran
Pyllula
Paieval
Anicrl

And heres a random sample of "names" generated by the transition probabilities for girls in 1913:

Lbeana
Peved
Math
Bysenen
Viel
Lelinen
Jabbesinn
Mabes
Drana
Lystha

The feeling I get looking at these is that they don't seem particularly gendered, even though there are clear gendered name trends. They don't even seem like they're from different times from each other. A lot of them aren't even orthographically valid. I don't know how they'd perform on "name likeness" tasks, but I don't even know what the point of doing such a task would be, since the Markov process has already failed at being a good model of names.

Maybe there's a lesson to be learned from the Markov process' failure to model names well, but for me it wound up being a silly diversion.

Update: I've now updated the package to generate names on bigram -> character transition probabilities. The generate_n_names2() function generates things that look more like names. It's kind of fun!

10 generated girl names based on 2010 data:

Rookenn.
Lilein.
Hayla.
Dailee.
Bri.
Samila.
Abeleyla.
Eline.
An.
Rese.

10 generated boy names based on 2010 data:

Briah.
Dason.
Jul.
Messan.
Kiah.
Jax.
Se.
Frayden.
Dencorber.
Gel.

America's Ugliest Accent: Something's ugly alright.

noreply@blogger.com (JoFrhwld) — Wed, 01 Oct 2014 15:28:00 +0000

I should really blog more often, instead of just when I feel compelled to slap down some nonsense, because the general tone of Val Systems turns towards scolding and away from my genuine positive passion for linguistics. That said, guess what I'm doing in this post!

If you read past the headline, it gets even worse. I won't always reply to examples I find of gross linguistic discrimination like this, because if I did it'd be a full time job. But I noticed that in the introduction they'd linked to a New York Times column that references a paper that I co-authored on the Philadelphia dialect. I didn't think the NYT column was appropriately respectful, and I said so on Language Log at the time.

The NYT columnist wasn't too happy about what I said, but I feel that I have an ethical obligation to the people who invite us into their homes and are generous with their time and stories, to provide them with a vigorous public defense if their communities and the way they speak are ridiculed as a result. Moreover, language shaming pieces like this Gawker tournament only poison the waters for future sociolinguistic research, especially if our names as researchers are attached onto them in some way.

And as I was writing up some notes for this response, and followed more links from the Gawker pieces, I was really shocked by how many articles they've linked to that are popular writeups of sociolinguistic research, usually including interviews with one or more sociolinguists! It's like half my facebook friends list in there! It feels so defeating to see these generally positive articles and interviews utilized to prop up an exercise as ugly and mean spirited as this one.

But what's the harm...?

Anticipating some reactions to this post, no, I'm not some grey humorless lump. But just because something is framed as a game doesn't make it fun, and it doesn't make it funny. For example, take Gawker's paragraph about New Orleans:

New Orleans is a steaming, fetid stew of aural bile, home to everything from the deep Cajun bayou accent to the Yat dialect, which derives from Irish, French, German, and even Italian into one completely incomprehensible mess. You need only watch this clip on the number of ways residents pronounce the city's name and neighborhoods and read this excellent article on the hodgepodge of New Orleans' accents to see how varied, and uniformly ugly, it all is.

lol?

Even if there was some intrinsic humor to that paragraph, instead of just raw nastiness, it would be important to reflect on who we're making fun of, and why. And on that count, I think the fella who compiled the Yat dictionary summarizes it well:

It's a working class language, probably, is what it amounts to.

And of course, that's what linguistic discrimination is really about. Maybe it's not always about class, but it's never really about language. It's about the kind of people who speak it. Predictably, the kinds of accents and languages which get dumped on the most, and get branded the "ugliest," always wind up being spoken by socially disadvantaged people. What exactly did this woman in particular do to deserve having a candid video of her slapped up on Gawker as an example of just how "ugly" the Chicago accent is? She works in a warehouse supermarket, that's what.

And this isn't a consequenceless game either. "America's Ugliest Accent Tournament" just puts a laughing face on a serious problem of discrimination that has economic and personal consequences for real people. To choose one example I'm familiar with, Anita Henderson did a study where she surveyed hiring managers in Philadelphia, playing them tapes of potential job applicants, and asked them to rate them on their job suitability. The topline summary from the abstract:

Those who sound Black are rated as less intelligent and ambitious and less favorably in job level.

In her textbook English with an Accent, Rosina Lippi-Green sums up my own opinion on the matter, but I've added some emphasis.

If as a nation we are agreed that it is not acceptable or good to discriminate on the grounds of skin color or ethnicity, gender or age, then by logical extension it is equally unacceptable to discriminate against language traits which are intimately linked to an individual's sense and expression of self.

How's this different from these other examples?

A few of the supporting links from the Gawker piece are personal websites that are called "How to talk [City]" or "The [Dialect] Dictionary," put together by enthusiastic speakers of the area themselves. They tend to have a self deprecating tone, so isn't that similar to the Ugliest Accent Tournament? It sure as hell isn't! First of all, even if those personal sites do have a poking fun tone, the fact is the dialect must be important to the person putting together the site, or else they wouldn't have spent the time documenting it! Their self deprecating tone could either be due to the general difficulty of expressing seriously how important a topic is to you, or to their internalized linguistic insecurities driven by things like America's Ugliest Accent Tournament. Moreover, Gawker is a really large media organization, and should be taken to task if only due to their profile and influence.

Sociolinguists ask people what they think about accents and dialects too. It's a subfield sometimes called Perceptual Dialectology. Isn't that kind of the same? Don't even start! A goal of sociolinguists is to understand the social landscape of language as well as we can, and that includes people's sometimes crummy attitudes about it. But if we have a goal, it's to critique those attitudes, not revel in them in some kind of user engagement experiment so that we can go cash in our pageviews with advertisers.

The Fundamental Sociolinguistic Outlook

Real quick, let's contrast the overall tone of the America's Ugliest Accent Tournament with what I would call the fundamental sociolinguistic outlook on speakers. I think Bill Labov summed it up nicely at the end of his 2009 Haskins Prize lecture (go listen to it if you haven't already).

Versus Gawker

No matter who you are, you all sound disgusting.

What to do about it?

I for one will be writing a polite e-mail to Gawker asking them to remove the link that references my research, and to avoid linking to anything that references my research in the future. I'd encourage anyone else whose research they mentioned to do the same.

The new iOS Health app is disappointing

noreply@blogger.com (JoFrhwld) — Thu, 25 Sep 2014 16:02:00 +0000

I've been getting into using a few different health tracking apps, and have been getting tired of needing to punch the same data into 3 different places every time I step on a scale. So, I was reasonably excited about the new Health app in iOS8, which would act as one central repository for this information that the individual apps could pull from. The fact that release of the HealthKit API has been delayed, meaning my 3 different health apps can't access the Health data yet, is disappointing, but I'm pretty patient about these things.

However, the Health app itself is really disappointing all on its own. It is not a success of data reporting or visualization. For example, here is what the record of the number of steps I've taken each day for the past month looks like.

So, riddle me this: How many steps did I take yesterday? What was the date that I took the most steps? What day of the week was that weird dip? Not only are answers to basic questions like these not "glanceable," they are totally inaccessible. There is, in fact, no way within the Health app to find these answers, but back to that later.

Let's get a bit more detailed. What is the range of the y-axis. It looks like the bottom horizontal like corresponds to 1,500 steps. That's already a questionable data reporting decision. It should probably correspond to 0 steps. How about the top of the y-axis range? The top horizontal like looks like it corresponds to 13,951 steps, but I'm actually pretty sure that is the maximum number of steps in this data. But the maximum data point doesn't touch the top line?

But let's talk about how Apple really failed to meet baseline expectations with these graphs. When I realized I couldn't read the data precisely off the graph, my first instinct was to drag my finger across the line, assuming that more detailed contextual data would pop up. Sort of like how this Google Ngrams graph works. It should even work on mobile if you tap on it. Or, take this excellent bit of interactive visualization from the New York Times Upshot blog. Or any line graph out there with any bit of polish. Users are more or less trained by this point that hovering over line graphs activates some kind of additional contextual information, whether it's more detailed labeling, brushing, or like that NYT visualization, additional graphs! So you might expect that on the baddest touch screen device ever in the world (as Apple would have us believe), there's going to be some wild and crazy touch interaction, pinch-to-zoom pizzazz. Or at least it might have the same baseline functionality as some silly web widget that I can embed in my blog.

No such luck, and the data viz nerd in me sees this as one of the biggest missed opportunities I've seen in a while. It is just a static image, with some minimal transition animations when you switch between different time scales. If you tap on the graph, you get taken to the raw data, which looks like this.

As far as I can tell, this is the really raw data offered up by the motion co-processor. Ludicrously, you can select and delete any individual bout of steps. So, if I felt that actually, one of the 8 groups of steps logged all in the minute of 10:08 AM was inaccurate, I could delete it!

What really frustrates me about the fact that I can see this data is that I can't touch it. Data at this granularity is pointless other than to show off the fact that there's a lot of it. It needs to a little bit aggregated before it gets interesting. And the fact that I don't like the Health visualizations as it is, I'd really go to town on this raw data. But conspicuously absent here is any export utility. I can look at, but not touch my own data. I guess I also couldn't access the data before iOS8, but they didn't waggle it tantalizingly in front of my nose like this!

So sure, maybe someone will make a third party app that will access the data from the HealthKit API and allow me to export it from there. As if what I'm really dying to do is clutter up my phone with an inevitably junky ad riddled app that contributes functionality that really should've been there in the first place.

To sum up, the static figures are poorly designed and minimally informative, but static figures are hardly what I would expect from a corporate entity like Apple anyway. On top of that, waving this raw data in my face is equal parts useless and infuriating.

Baby Naming Trends: Now With More Linguistics!

noreply@blogger.com (JoFrhwld) — Sun, 13 Apr 2014 14:25:00 +0000

This animated graph about the rise in boys names ending in <n> has been making the rounds lately.

It comes from this blog post by David Taylor.

It's a really cool graph, but then, I tend to find analysis of baby names a bit frustrating because they almost always rely strictly on the written, or orthographic, forms of the names. It's not that the way people spell their children's names doesn't matter, but it's half of the puzzle. For example, I'm named after my grandfather. He was German (more specifically, a Donauschwob), so he spelled his name <Josef>, and pronounced the initial sound like <y>, which in the IPA is /j/. When naming me, my parents had a whole bunch of options. Would the pronounce my name like my grandfather did, or like most English speakers would? And how would they spell it? They wound up settling on the English pronunciation, and the German spelling. I've made a little diagram displaying a very partial set of options my parents had in choosing my name.

And of course, Sarah Jessica Parker played a woman named /sændi/ who spelled it <SanDeE☆> in Steve Martin's LA Story, so clearly the spelling of proper names is an important expressive dimension, but still just half the picture.

So, I decided to look at a bit more at popular linguistic structures in baby names. Hadley Wickham has already compiled the top 1000 baby names in the US per year since 1880 (https://github.com/hadley/data-baby-names), and Kyle Gorman has a nice python module that syllabifies CMU dictionary entries (https://github.com/kylebgorman/syllabify). So I put together some sloppy code to analyze it (https://github.com/JoFrhwld/names). The biggest weakness to my approach is the number of names which are not to be found in the CMU dictionary. 2525 out of the total 6782 names in the data (about 40%) aren't in CMU, so this post should be understood as being for entertainment purposes only.

One other thing that bugged me about the name final <n> plot is that it seemed kind of arbitrary to focus on the final letter of the name. I suspect that it's a real trend that people noticed eyeballing lists of names, but that it wasn't compared against other kinds of trends. I went ahead and labeled name initial and name final syllables, codas, onsets and rhymes as being special, but I'm not going to single them out.

Kicking things off, there's a graph of popular syllables between 1880 and 2008. To be included in the graph, a syllable had to be in the top 3 most popular in any given year. The y-axis is how many times more frequent the syllable is than if syllable selection were random. It's not frequency rated, that is, this is just the distribution over names that have that syllable, not babies.

It's a bit chaotic, I know. It's a time like this that I wish I'd learned a little JavaScript so I could make an interactive version with brushing. Here's another version where each syllable gets a facet. They're ordered by their decreasing maximum ratio.

It looks like at the syllable level, name final /nə/ and /li/ for girls are both long time favorites, as well as more popular syllables than any boy's name final /n/ syllable. The most popular boy's name final /n/ syllable looks like it's always been /tən/, but maybe it's flagging a bit compared to the recent surges in /sən/ and /dən/. It also looks like popularity in syllables is pretty evenly split between name initial and name final syllables. For both boys and girls, some kind of initial between /e/ ~ /ɛ/ ~ /æ/ is pretty popular, but I can't be sure what's going on there, because the CMU dictionary has the same entry for both <Aaron> and <Erin>.

But maybe the reason boy's name final /n/ isn't shining through like you might expect is because of phonological reasons. A boy's name ending in a word final syllabic /n/ is necessarily going to pull the preceding consonant into the syllable with it. Looking at the plot above, it's not likely that the preceding consonant is totally random either, cause we've only got /t, s, d/ (all coronals) and vowels preceding the /n/. But for the hell of it, here's the same kind of plot as the ones above, but this time with syllable rhymes.

There's a lot less volatility in the rhymes data, probably because there's fewer different kinds of syllable rhymes. Complex rhymes don't seem to be that popular ever. We've mostly got vowels from open syllables, and syllabic consonants. At any rate, the popularity of name final /n/ for boys is pretty clear, taking over from /i/ (from names like Billy and Jonny). The boy's trend towards name final /n/ seems to be about on par with the trend for girls names to end in /ə/.

I'd like to play around with this data a bit more if I get some time. It occurred to me that you could come up with a few different ways of generating popular names from different eras by randomly sampling popular syllables, or by estimating transition probabilities between syllables and going on a random walk that way.

All my code and the data are up on github, if anyone else wants to play around with it: https://github.com/JoFrhwld/names.

When did Americans start spelling it "color"?

noreply@blogger.com (JoFrhwld) — Thu, 23 Jan 2014 12:51:00 +0000

Usually I wouldn't apologize for a lapse in posting, since I think an obligation to apologize acts as block to actually making another post. But in this case, my hiatus is (loosely) related to the topic.

In September, I defended my dissertation (check it out here if you're so inclined), then hopped on a plane and immigrated to Scotland to start a job as a lecturer in Sociolinguistics at the University of Edinburgh!

The sun was in my eyes. I was feeling very excited.

It's been a fun experience exploring the new cultural landscape. As is usually the case, the differences are more noticeable and surprising than the similarities, but I'm managing more or less. I'm crossing streets with confidence that I know which way the cars are coming from, participating in rounds at pubs (though I'm still not sure I'm doing that right), and thanking the bus driver, which is nice enough to do in the States, is apparently more obligatory here.

One thing I was told by another recent US immigrant to the UK wis that that students appreciate attempts at accommodation to British spelling. So, I'm giving that a shot too, although I'm sure I slip up a bunch, since I'm a very poor copy editor. However, out of passing curiosity, I decided to look at the historical trends in these spelling differences in the Google ngrams, and the patterns seemed interesting enough to warrant this blog post.

First, looking at the American English data for <color> and <colour> there's a very nice and clear cross over from to around 1845, which is about 20 years after Webster's 1828 dictionary, which according to Wikipedia is what we have to blame these differences on. Once you start trying to add more <-or ~ -our> to the graph, it gets chaotic fast, so instead of plotting out each word, I'll plot out the percent of <-or> spelling using Google ngram's handy arithmetic functions. (I've also included color/color, just to anchor the top of the range at 100%) I don't think I should have been, but I was a bit surprised with the uniformity with which the <-or> spelling replaced the <-our> spelling across all of these words. The trends seems to kick off around the 1820s, consistent with blaming it on Webster's dictionary, and increased till reaching its plateau around 1860.

But of course, <-or ~ -our> spelling isn't the only difference between British and American systems. The next set of consistent spelling differences involve <-er ~ -re>. Here are those words plotted out, with <color ~ colour> and <humor ~ humour>left in there as a representative items of the <-or ~ -our> set. So, it seems like there is a similar uniformity within the <-er ~ -re> words (maybe saber and theater are lagging behind) but the <-re→-er> replacement is offset from the <-our→-or> by about 60 years or so.

Of course, I shouldn't have been surprised, because I know how a little bit about language change, but it was fun to see this thing that I think of being a uniform "American Spelling" is actually the result of multiple changes that didn't happen all at the same time.

Just for fun, I took a look at what these patterns look like in British English. So, it looks like there might be a bit of a creep of American spellings into British English, but interestingly, the particular alternations aren't differentiated. So while the end product of "American Spelling" appears to be the result of an accumulation of different changes, the borrowing of American spelling into British English is being done holistically.

The relationship between Kanye, Rap and African American English

noreply@blogger.com (JoFrhwld) — Fri, 21 Jun 2013 03:53:00 +0000

This started out as an update to my post on Kanye West's song "I am a God," but wound up being nearly as long as the post itself, so I'm separating it out. But look over that first, or maybe keep it open in a separate tab.

A concern has been conveyed to me that I may have been equating rap as a lyrical form, African American English, and Kanye West in a problematic way. I wasn't really clear about my assumptions about how these three things are related, so I'll try to clarify.

So first, I definitely don't want to imply that the conventions of what is possible and not possible in rap is equivalent to the grammar of African American English. Rap is strongly identified as an African American art form such that people frequently malign hip hop as code for AAE, but as a linguist I know better than to draw similar equivalencies. As a lyrical and musical form, rap has its own conventions which aren't the same as AAE grammar. This must be the case, because speakers of other dialects and languages can produce songs which are clearly identifiable as rap!

I am assuming that Kanye West is a speaker of AAE as it spoken in Georgia, and that his phonology, as he acquired it, generates a bunch of representations. That's why I tied in the Labov, Cohen & Robins (1968) reference, to try to emphasize that these ∅ coda Go(d)s weren't just a quirk of Kanye, but rather reflective of larger dialectal trends in which Kanye is a participant. What does it matter? It's just more interesting if the reasoning I proposed here could generalize beyond just Kanye.

Next, I assume that Kanye has a personal filter, partially due to his personal taste, partially due to the conventions of rap, whereby he decides whether two words work as a rhyme. My reasoning is that if we understand Kanye's filter, and can see what comes out the other end, then we can make some assumptions about what went into it in the first place. Importantly, Kanye's rhyme filter is not AAE. In AAE, the set of words {God, Go(d), massage, ménage, garage, restaurant, croissants} definitely aren't perfect rhymes, but they all passed through Kanye's filter as rhyming equivalent.

My reasoning is as follows. All of these words passed some metric of Kanye's rhyming filter as being equivalent enough. The zero coda variant "Go(d)" is a product of Kanye's AAE phonology. If we can figure out what Kanye's filter is, then we can know something about the phonological status of "Go(d)", which can tell us something about this feature of AAE phonology. Excluding "Go(d)", all of the final syllables of the words in this rhyming scheme have 1) a low-back vowel 2) a coronal obstruent. The coronal obstruents vary quite a bit in the sub-coronal place of articulation, their manner, and in their complexity. So, I want to conclude that Kanye's filter requires matching on the vowel quality, and the major place of articulation of the coda, but not its manner or complexity.

The zero coda Go(d) is an outlier in this pattern, unless we conclude that the missing /d/ actually counts as being there. Whether or not the missing /d/ counts as being there has more to do with Kanye's phonology than his rhyming filter, and since I'm assuming Kanye's phonology comes form AAE, this could also be a property of the phonology of many AAE speakers. So, I want to conclude that it is probable that for many AAE speakers, when they produce zero coda Go(d), the /d/ still counts as being there.

Now how gone is the /d/? That's where the phonetics comes in, and the answer seems to be "very."

This is just a very general view of how to use rhyming verse to figure out something about the phonology of any language or dialect. A speaker of some language has some phonology which generates forms, and then they have a rhyming filter to see what works. By working out what the properties of the filter might be, you can try to reconstruct what the properties of the phonology is.

Kanye's Codas and Vowels

noreply@blogger.com (JoFrhwld) — Thu, 20 Jun 2013 17:36:00 +0000

This is just a quicky about Kanye's new song I am a God, which he collaborated on with God. The opening lyrics are:

I am a God
I am a God
I am a God
Hurry up with my damn massage
Hurry up with my damn ménage
Get the Porsche out the damn garage
I am a God
Even though I'm a man of God
My whole life in the hands of God
So y'all better quit playing with God

Kanye pronounces the final syllable of "massage" as [ɑʤ], which caught my ear so I went back and listened more closely. By my coding, this is how Kanye produces the final consonants of each of these lines.

God: [d]
God: [d]
God: [d]
massage: [ʤ]
ménage: [ʒ]
garage: [ʒ]
God: ∅
God: ∅
God: ∅
God: ∅

Those ∅ codas mean that the /d/ was just absent. I believe AAVE deletes simple codas more often than other dialects. I don't know more recent numbers, but Labov, Cohen & Robins (1968) reported that in Harlem it happened somewhere between 5% and 20% of the times for a word like "God", depending on the context.

The first thing that occurred to me was that for Kanye, [gɑ:] is rhyming equivalent to [gɑd]. That might be an interesting piece of evidence for the phonological status of these zero codas. That is, they really count as having a /d/ in them, even though it's not pronounced. Then again, I don't know if Kanye would be willing to rhyme "spa" with "God", meaning he doesn't really care about the content of coda, as long as the vowels match. Later on in the song, though, he pulls in "restaurant" and "croissant" into the same rhyming scheme, making the set of codas which are rhyming equivalent {d, ʤ, ʒ, nt, ∅}, all apical obstruents except for the zero codas. It seems like Kanye doesn't care about coda complexity, manner, or subcoronal place of articulation, but the major place of articulation does seem to matter, so I'm going to say that those ∅-codas actually count as apical obstruents for Kanye.

But, Joel Wallenberg asked me an interesting question on Twitter.

@JoFrhwld Does the vowel transition towards closure in the 0 ones, even though there's no real closure?
— Joel Wallenberg (@quakerscientist) June 20, 2013

Well, there's only one way to find out! Break out Praat!

Here are the first three God's, which I heard with a [d]. They all have pretty clear formant transitions (marked by "tr"), especially the third one. (click for larger)

God 1

God 2

God 3

And here are the last 4 Gods, which I heard without a [d]. God 4 does actually have a pretty clear formant transition for an apical closure, and God 6 has something subtle going on that I'm not sure about. It looks less like a formant transition to an apical closure, and more like a return to a neutral vowel, but I've marked it "tr?" just in case. Gods 5 and 7, though, clearly don't have formant transitions to a following /d/.

God 4

God 5

God 6

God 7

So, for at least 2 out of the 4 zero coda Gods, the /d/ was really really gone.

One more interesting thing to me was that except for Gods 1 and 7, after reaching their peak F1 and F2 after the transition from [g], all of these vowels have a very steadily declining F1 and F2. In fact, I'd transcribe God 2 as being something like [gaɔ̯d]. I guess for Kanye the vowel in God is definitely long and in-gliding.

Update:

A concern has been conveyed to me that I may have been equating rap as a lyrical form, African American English, and Kanye West in a problematic way. I wasn't really clear about my assumptions about how these three things are related. I tried to clarify in this post.

Socioeconomic Status and College Enrollment

noreply@blogger.com (JoFrhwld) — Wed, 22 May 2013 21:58:00 +0000

There was an interesting post over at Sociological Images about the relationship between socioeconomic status and high school students likelihood of even applying to highly selective colleges.
The research cited there by Alexandria Radford focused on high school valedictorians, which I think is a pretty cool way to isolate the effect of socioeconomic class. Sociological Images posted this figure from the paper (abstract? proposal?) which shows what proportion of valedictorians wind up in each category with regards to highly selective colleges

They applied to at least one highly selective school.
They were accepted to at least one highly selective school.
They enrolled in a highly selective school.

Radford attributes the large gap between High SES valedictorians and Middle and Low SES valedictorians to reduced familiarity with highly selective schools as institutions and their admissions process. That seems to ring true to me, but I was curious how SES affects each step in this process. The graph above doesn't really tell you what proportion of Low SES students who actually applied to highly selective schools were admitted, or at least I don't believe it does. It says 42% of Low SES valedictorians were admitted to a highly selective school, but you have to apply to be admitted, and only 50% of Low SES valedictorians applied in the first place. That means that 42%/50% = 84% of Low SES students who applied were admitted. The same goes for enrollment. You have to be admitted to be enrolled, meaning the 32%/42%=76% of Low SES students who were admitted to a highly selective school also enrolled. Here's a rejiggered plot I made based on the data in the plot above.

So, it looks like to highly selective colleges' credit, the proportion of high school valedictorians who sent in applications who were then admitted is just about flat across the socioeconomic scale. 85% of High SES valedictorians who applied were admitted, and 84% of Low SES valedictorians who applied were admitted. Things look really different for enrollment though. High and Low SES students have very similar enrollment yield, (79% and 76%, respectively), but Middle SES students have much lower enrollment yield (59.6%).

I can only speculate why Middle SES students have such a lower enrollment yield than Low and High SES students. One possibility is that their family's income is just a little bit too high to qualify for needs-based scholarships. Another might be that the perceived cost-benefit ratio for Middle SES students is higher than for Low SES students.

However, I think this plot lends credence to Radford's hypothesis that reduced information and familiarity leads to reduced Low SES application rates. For Low SES valedictorians, their application rates are totally disproportionate to the probability that they will be accepted to the highly selective schools, and to the probability that they will actually attend those schools if they are accepted.

Update: I meant to include the R code for the data and plot.

Update 2:
Of course, admission is not even half or a quarter of the battle. I don't know much about the The High School Valedictorian Project mentioned in the study, but I hope they keep tabs on the Low SES students admitted to the highly selective schools. I know the culture shock I experienced going from a good diocesan high school to an Ivy League university was pretty huge, but I don't know if it was disproportionately huge than it would have been going from any old high school to any old college, or what effect it might have on academic performance.

I Support Inclusive Scouting

noreply@blogger.com (JoFrhwld) — Tue, 21 May 2013 23:50:00 +0000

This week, the Boy Scouts of America's National Council will be meeting to vote on whether the Boy Scouts should include the following statement in its membership policy for youths.

"No youth may be denied membership in the Boy Scouts of America on the basis of sexual orientation or preference alone."

I thought I would take a brief moment to publicly state my support for a more inclusive scouting, and one that will hopefully eventually welcome gay and lesbian adult leaders, and non-religious youths and leaders.

I am bothering to speak out about inclusive scouting because I believe that scouting is an over all positive experience that has great value for the youths involved in it. I outlined some of the benefits I felt that scouting had brought to my life in the letter I enclosed with my Eagle Scout award when I returned it in protest of BSA's exclusionary policies. For example, it was in the scouts that I had my first experiences with organizing with peers to reach a goal. The skills I learned there, like learning how to clearly express my vision of what should be done, how to set aside that vision in favor of ideas offered by others, and how to delegate responsibilities have been invaluable for me in my professional life. I also learned the value of living a life of principle, which is why I'm writing this blog post. I was also able to bond with my father, who is also an Eagle Scout, and my brother. I don't believe that there is any good reason to deny any youth or parent any of these experiences on the basis of their sexual orientation, gender identity, or metaphysical beliefs.

On this specific issue, I would like to preempt most objections to allowing gay youths into the scouts by coming back to the supreme court ruling that BSA won. The most most common concern I've heard voiced about allowing gay youths into the scouts revolves around who will sleep in which tent on a camping trip. Second and third most common are the discomfort other youths and sponsoring organizations may have with gay youths being in their troop.

But the majority opinion in Boy Scouts of America v. Dale made no mention of the logistics of sleeping arrangements, nor discomfort of individuals or sponsoring organizations when they decided that BSA has constitutional right to exclude gay youths. Rather, it was argued that the Boy Scouts of America had an expressive policy condemning homosexuality, that the Boy Scouts "teach that homosexual conduct is not morally straight." The constitutional right to free association that BSA won was not won on the basis of something so trivial as how tricky sleeping arrangements will be, but rather by arguing that they have a core expressive policy that being gay is wrong. When voting on whether to adopt the new membership policy, I hope that National Council members have in mind that they are voting on the principle of whether or not being gay is wrong, and that other logistical concerns are secondary.

Regarding those logistical concerns I would just briefly say that the concern I have about openly gay youths in shared tents is the real threat of violence towards the gay youths. That is the reality we are living in. Scoutmasters will have to emphasize to their troops that violence towards each other is not ok, and that we should have to treat all people with human decency, a value I should hope is well within the expressive mission of BSA. As for the discomfort of youths and sponsoring organizations at having gay youths in their troops, I would just say that one of the most valuable experiences I had in the scouts was being in a troop with some people who I would have rather not have been. Not any class of people, mind you, just individuals who at first brush I probably wouldn't have volunteered to be friends with. We don't get to choose who lives in the world with us, but we have to live and work with them every day, and hopefully we can also learn to see the value in them. I would say it is doubly important for youths brought up with religious beliefs that strongly condemn gays and lesbians to be friends with and work with gays and lesbians their own age. They are growing up in a world where gays and lesbians (fortunately) feel more comfortable being out about their orientation, and they will be forced to reconcile their religious beliefs with the fact that they will have gay and lesbian peers as adults, and they will have to do so with grace and aplomb.

I know that I haven't addressed all of the concerns people have raised about allowing gays into the Boy Scouts. For example, some people have had fairly vile things to say about gays and pedophiles which I won't bother myself with here. And I won't bother myself with any other arguments either, since most of them seem to have the property that once you provide a counter argument, a new argument can be freely generated. I think this is because these concerns about camping in tents and so on are transparently secondary to the core contention about whether being gay is wrong, a proposition which has become impolite to say itself. Hopefully, the National Council adopts the new membership policy.

A difference between men and women.

noreply@blogger.com (JoFrhwld) — Sun, 17 Feb 2013 23:08:00 +0000

This post was originally going to be a lot more mathy, with a bit of explanation about the source-filter model of speech production with an aside about dead dog heads mounted on compressed air tanks thrown in there, and a whole description of my methods, but I felt like I was sort of burying the lede there. Instead, I'm focusing more on how people are interested in magnifying the difference between men and women.

It started off with me estimating the vocal tract lengths of the speakers in the Philadelphia Neighborhood Corpus. Given sufficient acoustic data from a speaker, and making some simplifying assumptions, and taking into account the acoustic theory of speech, you can roughly estimate how long a person's vocal tract (meaning distance from vocal cords to lips) is. I went ahead and did this for the speakers in the PNC, and plotted the results over age.

Pretty cool, right? There's nothing especially earth shattering here. It's known that men, on average, have longer vocal tracts than women. I was a little bit surprised by how late in age the bend in the growth of vocal tracts were.

Here's the density distribution of vocal tract lengths for everyone over 25 in the corpus.

That's a pretty big effect size. Mark Liberman has recently posted about the importance of reporting effect sizes. He was focusing on how even though people are really obsessed with cognitive differences between men and women, the distributions of men and women are almost always highly overlapping.

Following Mark on this, I went ahead and calculated Cohen's-d for these VTL estimates.
So, 1.71 is a fairly large Cohen's-d effect size. I had heard that the difference in vocal tract length between men and women was disproportionately large given just body size differences. I managed to find some data on American male/female height differences, but the effect size is not impressively smaller than the VTL effect size (1.64, about 95% the VTL effect size).

Compared to the effect that Mark was looking at (science test scores), these effect sizes are enormous. The effect size of height between men and women is about 23 times larger than the science test score differences which warranted a writeup in the New York Times.

Yet, still not big enough.

As I was thinking about how height difference is perhaps one of the largest statistical differences between men and women, it also struck me how often it is still not big enough for social purposes. Sociological Images has a good blog post about how even though Prince Charles was about the same height, if not shorter than Princess Diana, in posed pictures he was posed to look much taller than her. Here's an example of them on a postage stamp:

And in another post, they provide this picture of a reporter being comically boosted to appear taller than the woman he's interviewing.

My take away point is that when it comes to socially constructing large and inherent differences between men and women, even the largest statistical difference there is out there is still not good enough for people, and needs to be augmented and supported. Then take into account that most other psychological and cognitive differences have drastically smaller effect sizes, and it really brings into focus how the emphasis on gender differences must draw almost all of its energy from social motivations, rather than from evidence or data or facts.

I recommend Lexicon Valley

noreply@blogger.com (JoFrhwld) — Thu, 07 Feb 2013 19:50:00 +0000

Perhaps the most frustrating thing about being a linguist is the enormous gap among educated people about how little they actually know about language, and how confident they are that they know a lot about language. If you keep up with this blog, I spend a lot of time venting this frustration here (etc. etc. etc.).

But I didn't start blogging in order to complain about how other people are getting it wrong. I started blogging to have an informal outlet for passion for linguistics! I've been a little concerned about the negative tone of a few of my recent posts, so here's a more positive one.

But... it does start off with a complaint. At the LSA this year, David Pesetsky's plenary focused on the failure of linguistics (and more specifically, generative linguistics) to penetrate the popular science press. Instead, stories about physicists discovering the most common English word is "the," and psychologists arguing that structure of language is really words like beads on a string get a lot more play. At the Q&A, Ray Jackendoff made the point that there is a folk linguistics that is intricately tied up in social politics that acts as a major roadblock to the popular advancement of real linguistic research. I've said similar things before.

What is to be done about this state of affairs is the topic of another blog post. Right now, I'd like to bring attention to a bright light of potential linguistics popularization.

Lexicon Valley

Lexicon Valley is a podcast hosted by Slate. I've been listening to it off and on since it started, and I have to say I've always enjoyed it. The hosts play two roles in a dialectic. Mike Vuolo is the patient intellectual, and I've always been impressed by the background research he's done. Bob Garfield is the voice of the untutored establishment, and, well, I think that description adequately sums up my opinion of what he brings to the show. It's actually an important role he plays, because without a vocal foil, Vuolo's research would lie rather flat. It's also important for the cause of linguists to have people hear brash knee jerk reactions rebuked by careful research.

They have covered a few topics I know a little bit about, and I've always started listening to each show bracing myself for frustration and disappointment. It's a learned reaction I have from every other discussion of language in popular media. But Lexicon Valley usually carries through for me. They've done great shows on African American English, grammatical gender, and the English epicene pronoun, speaking to actual linguists in each case, and most recently they've just done a really good portrayal of Labov's department store study (Part 1,Part 2).

They did catch a lot of flack recently for their show on creaky voice. I was so nervous when I started listening to it, because the recent coverage creaky voice has gotten has been worse than terrible. Per usual, though, Vuolo's research and discussion were excellent. Garfield, on the other hand, spouted some really negative attitudes, and I think he deserves every criticism of sexism that he got. Even within the dialectic of the show, Garfield brought a net negative contribution that time round. On the subsequent show, though, Vuolo read out some pretty harsh commentary about Garfield. Garfield offered a nonpology (something about how he can't be sexist, he has daughters), but it was good to have some of the criticism read out loud.

On average, modulo Garfield's frustrating attitudes, I would highly recommend the podcast, and would recommend recommending the podcast.

Could it be better?

While I think Lexicon Valley has done some great work so far, I don't think it has yet provided coverage of linguistics in quite the way Pesetsky dreams of. So far, they've mostly covered topics that are reactive to popular gripes or misconceptions about language. In some respect, it'd be hard for them to do otherwise, because the popular understanding of language science is far below that of almost any natural science, or so it seems from this angle.

I hope, though, that they might find a way to approach linguistic topics which are not just reactive. Just addressing the idea that there are functional elements which have no phonological realization would be enormous. Garfield could play the skeptic, believing that what you see is what you get.

So linguists, listen in, get a feel for the show, and maybe if you have a topic which could be nicely formatted into a 20 minute conversation, send it in to them!

Does language "cool"?

noreply@blogger.com (JoFrhwld) — Sun, 03 Feb 2013 23:36:00 +0000

A few months ago, I posted about how I was relatively unimpressed by a paper arguing that the observed Zipfian distribution of words in a corpus is due to "preferential attachment" aka the Matthew Effect aka the rich get richer. The author of that paper is apparently also a co-author of a paper called "Languages cool as they expand: Allometric scaling and the decreasing need for new words." The writeup in Inside Science summarizes it like this:

[A] recent analysis has found that as a language grows over time, it becomes more set in its ways. New words are always being added, according to this study, but few become widely used and part of the standard vocabulary.

My linguist hackles immediately raised at this statement, and that's because there is a large and fundamental difference between what a linguist understands the term "language" to refer to, and what the authors of the column and paper understand it to refer to. What the physicists and the reporter mean by "language" is roughly "a set of words," and in the context of the paper, they almost seem to mean "the set of words which have been published."

This "language is words" axiom is part of most people's folk linguistics that we have to train people out of when they take Intro to Linguistics. That's why it's a little hard to take the work of these physicists seriously at first glance. It is as if they were trying to write a serious paper on biological evolution with the assumption that traits acquired by an organism during its life were inheritable.

But there is an aspect of linguistic knowledge relating to the set of words and morphemes a speaker knows, which linguists call the "lexicon". So, I'll just go ahead and reread the paper mentally replacing each instance of "language" with "lexicon" in order to get through it.

Overall Thoughts

This paper seems to be a relatively competent (modulo Mark Liberman's concerns about OCR errors) description of the statistical properties of large corpora. But that's really as far as I think any of the claims can go. I am totally unconvinced that their results shed any light on language change, development, evolution, etc. I'm not even sure that the simplest statement that "the lexicon of languages has grown over the past 200 years" can be supported by the results reported.

The key problem that I see with the paper is the conflation of "new to the corpus" and "new to the lexicon." Here's how the problem of sampling language was describe to me, and I believe it goes back to Good (1953) and is key to Good-Turing Smoothing. Say you are a entomologist working in a rain forest, trying to make a survey of insect life. You put out your net for a night to collect a sample, then count up all the species in your net. Some bug species are going to be a lot more frequent than others. You'll have some species that show up many times in the net, but even more species will show up in the net with only one member. Now, let's say that you come back to the same rain forest two years later, and repeat the sample. You are nearly guaranteed to observe new species in your net this time around, but the key question is whether they are just new to the net, or are they new to the rain forest. If they're new to the rain forest, did they migrate in, or are they hybrids of two other species, or has a species you saw previously evolved really rapidly so that you're seeing it as different now?

These are really interesting and important questions for our entomologist to answer, but you cannot arrive at a definitive answer based simply on the fact that this new species has now showed up in your net. In fact, depending on a few factors, the answer with the highest probability is that the new species is simply new to your net. The Good-Turing estimate of the probability that the very next bug you catch will be a new species is that it's roughly equivalent to the proportion of bugs you've already caught that belong to a species you've only seen once.

The situation gets even more confusing if you come back to the same rain forest two years later with a net twice the size.

The paper has a figure plotting the increase in lexicon size over time. My first thought when I saw it was that it must be the case that the overall size of the corpus at each time point must also be going up. Coming back to the entomologist in the rain forest, the number of species in his net is merely a sample of how many species there are in forest. In the same exact way, the number of words in a lexicon can only be estimated by the words which people happened to write down. As you increase the size of the net, you're going to find more species which were already in the forest, but not in your net. As you increase the size of your corpus, you're going to find more words which were already in the lexicon, but not in the corpus.

Now, you need to add to this that at any given point in time, the true maximum number of possible words you could potentially observe in any given language is ∞. Yes, in fact, the whole reason language is interesting to study is because given a finite set of mental objects, and a finite set of operations to combine them, you can come up with an infinite set of stings, and that goes for words too, not just sentences. In 1951, "iPod" was a possible word of English, it just wasn't used, or at least not for the same purpose it is now.

Regarding the question of whether the "active" (as I'll call it) lexicons of languages have grown over the past 200 years, well, indeed, the overall number of printed words has also increased. Almost all of their results seem to have more to do with the technological development of publishing than it does with any other linguistic or cultural development. It is as if the entomologist said that over the past decade, the biodiversity in his rainforest has exploded, when really what's going on is his nets have been getting progressively larger.

Now, it might be the case that the active lexicon has grown more than would be expected given the increase in the size of the corpus year over year, but as far as I can tell, the authors did not try to estimate whether this was the case.

What about this cooling down?

The "cooling" effect referred to by the paper is the suggestion that as a language "grows" (which as I just said is dubious), the frequency with which particular words are used becomes more stable. Some words are more frequent than others, but words are less likely to move up and down in frequency over time/as the lexicon grows. Back to entomology, the suggestion is that as more species cram into a rainforest, each species is less likely to become more or less populous.

Again, though, the frequency, even relatively frequency, of a word in a corpus is merely an estimate of its true frequency. As the size of the corpus increases, so should the reliability of its frequency estimates, and we would predict decreasing volatility of those frequency estimates. The authors check for this, and find exactly this relationship between corpus size and frequency volatility, but I can't tell whether there was excess "cooling" left over. I wish they had said, "there was x proportion of cooling left unaccounted for by simply accounting for the size of the corpus," but I think this is perhaps another symptom of the assumption that the corpus=the lexicon=the language that I complained about before.

The Allure of Big Data

The reporter who wrote the Inside Science article did what it appears that the editors of Scientific Reports did not, asked a linguist to comment on the paper. Bill Kretzschmar was "underwhelmed," saying that most of these results are not new to linguists. I would take this as a word of warning about the allure of big data. The results discussed in this paper are not, by and large, new, but rather have never been done with data of this scale. But unfortunately, a fact which is already known does not get more interesting when it is reestablished with data 100 or 1000 times larger than before.

To take "Zombie Nouns" seriously, you must've had your brains eaten.

noreply@blogger.com (JoFrhwld) — Tue, 27 Nov 2012 19:58:00 +0000

At first, I didn't feel like blogging about the NYT Column on "Zombie Nouns" because I feel like I've been spending too much time being critical here, arguing against usage advice like this is futile, and I knew Mark Liberman would cover it. In fact, I drafted this post all the way back during the summer, and just let it sit. But now, I've seen the column, nearly verbatim, pop up on TED-Ed as a fully animated "lesson", which presumably means some educators are actually assigning it to classrooms of fertile and impressionable minds! It really can't pass without comment now.

Helen Sword says that you should avoid using nominalizations, which she calls "zombie nouns." They're nouns that have been made out of other parts of speech. To take one of her examples, calibrate + ion = calibration.

What is so wrong about nominalizations? Not exactly clear. She seems to take aim at unnecessarily jargonistic writing, which frequently contains novel coinings of words of all types, including nominalizations. So sure, being jargonistic to obscure your other intellectual shortcomings is not so good. But is it really, actually, the mere use of nominalizations that's doing the damage there?

She also seems to take a page out of the anti-passive voice book, saying, "it fails to tell us who is doing what," which just like the passive, is just not true. For example, in the sentence

My criticism of her column is a day late and a dollar short.

It's very clear who is doing what, even though I used a nominalization (in bold).

But on top of the half baked usage advice, there are some more reprehensible social attitudes being expressed. For example, she lists epistemology as a useful nominalization for expressing a complex idea, but heteronormativity as one only out of touch academics who are enchanted by jargon use. First off, I would not want to use epistemology as an example when explaining what nominalizations are. What's it derived from? Episteme? Episteme has a Wikipedia page, so I guess it's that. Which brings me to the next issue here. It's embarrassing for me to admit, but whenever someone says or writes epistemology, I have to go look it up on Wikipedia. How does using epistemology not count as being out of touch with how ordinary people speak? Heteronormativity, on the other hand, is pretty easy to wrap your mind around. From Wikipedia:

Heteronormativity is a term to describe any of a set of lifestyle norms that hold that people fall into distinct and complementary genders (man and woman) with natural roles in life. It also holds that heterosexuality is the normal sexual orientation, and states that sexual and marital relations are most (or only) fitting between a man and a woman. Consequently, a "heteronormative" view is one that involves alignment of biological sex, sexuality, gender identity, and gender roles.

That's a pretty complex idea. But you know what? It's pretty easy to decode most of that meaning from the word itself, at least, if you're vaguely familiar with the politics of the time. Hetero(sexual) + normative + ity. It seems to me that she's saying more about her position on sex and gender politics here than she is about usage advice.

But who is this person, and why is she writing an opinion column in the New York Times, and getting the full TED treatment? Just like everyone, she's selling something: the icing on the cake, and my reason for blogging about this at all. She has a book out called The Writer's Diet, which has an accompanying online Writer's Diet Test. No, it's not diet as in "food for thought and inspiration," like a Chicken Soup for the Writer's Soul. It's diet as in dieting as in "drop 20 lbs and get the six pack abs you always wanted." Just paste in a paragraph of your writing into the test, and it'll rate you along a five point scaled labeled:

lean

fit & trim

needs toning

flabby

heart attack territory

Ain't nothing like exploiting the collective dysmorphia of a nation to push your quarter-baked usage decrees. But in doing so, Sword actually clarifies the role that books like hers play. The analogy to the diet and weight loss industry is entirely apt. The dieting industry makes their money by sowing seeds of personal insecurity, then reaps their harvest with offers of unfounded, unscientific, and ultimately futile dieting pills, products, methods, 10 step plans, meals, regimes, books, magazines, etc.

I won't mince words. The NYT column and the TED-Ed video have the equivalent intellectual content of the magazines in the supermarket aisle promising you 5 super easy steps to trim your belly fat to get a sexy beach bod in time for the summer. And they serve the same purpose: to undermine the confidence of every-day folk, so that they may be taken advantage of by self-appointed gurus.

Creative Work

noreply@blogger.com (JoFrhwld) — Thu, 15 Nov 2012 19:14:00 +0000

Whenever I hear "creative" people describe their creative process, or more precisely their creative woes, I am always struck by the strong similarities to my own experiences trying to do science. I do consider myself as trying to do science.

Take, for example, this excellent statement on self-disappointment at the early stages of your career from Ira Glass.

Ira Glass on Storytelling from David Shiyang Liu on Vimeo.

This almost perfectly sums up how I felt about almost all of the early work I did in graduate school. I can't say that I've actually gotten to the point where the work I produce meets up with my my own personal standards, but it has been on an upward trend, and I'd say Ira Glass' advice is spot on. If you want to write good papers, just write a lot of papers, and if you want to be good at giving talks, give a lot of talks, preferably in a context where you feel comfortable being bad or mediocre.

That last bit, being comfortable with being bad is really reminiscent of things Brother Ali says in this interview.

Ill Doctrine: Brother Ali Meets the Little Hater from ANIMALNewYork.com on Vimeo.

There are a few things Brother Ali says that really resonate with me.

There was a moment where I was so stressed out. And I'm like, "Man, everything that I ever did that people liked, I just got lucky. I'm a fraud."

...

It's a weird weird thing to have what you create also be your livelihood. What we create is also our sense of self. What we create is also the way the world views us.

...

And so I start thinking about it. Ok, it's not that I'm blocked. It's not that I don't have anything to say. It's that I don't know how to say what I need to say. Or it's that I don't think that it's going to be received well. Or it's that the people that love me and have supported me and have, you know, gave me the little bit of freedom in my life that I have, I don't want to let them down and I don't want to hurt their feelings by saying what needs to be said.

I think almost all academics of any variety feel this way from time to time.

But I wonder if some people might not be surprised that I would feel so similarly to creative artists in the pursuit of my science, or that maybe take it as evidence that I what I do is not science. It is certainly doubted about Linguistics occasionally. But I think these people (probably strawmen) are mistaken in thinking that science is not a creative process. This was recognized by Max Weber in is 1918 essay "Science as a Vocation" (which I've blogged about before).

[I]nspiration plays no less a role in science than it does in the realm of art. It is a childish notion to think that a mathematician attains any scientifically valuable results by sitting at his desk with a ruler, calculating machines or other mechanical means. The mathematical imagination of a Weierstrass is naturally quite differently oriented in meaning and result than is the imagination of an artist, and differs basically in quality. But the psychological processes do not differ. Both are frenzy (in the sense of Plato's 'mania') and 'inspiration.'

He also suggests that the best science and the best art is produced by individuals devoted to the science and art for their own sake, rather than being driven by the express goal of producing something new, for the sake of novelty.

The distinction that Weber draws between art and science is that science is necessarily committed to the abandonment of old science. That is, art from the Renaissance is still, and always will be, art, but science from the same period is no longer science. It has been superseded by more recent developments.

Anyway, here's the song Brother Ali was talking about, which I'm sure almost all academics can identify with, except for the suicide ideation, hopefully.

Nate Silver vs.the baseline

noreply@blogger.com (JoFrhwld) — Wed, 07 Nov 2012 23:32:00 +0000

The 2012 election has been declared a victory for Nate Silver. As Rick Reilly said:

If Nate Silver told me it's going to rain marshmallows tomorrow, I'd stand outside with cups of hot chocolate.
— Rick Reilly (@ReillyRick) November 7, 2012

For me, as a data geek, this is nothing but good news. There's been a lot of talk about how Silver's high profile during the election could have broader effects on how every day people think about data and prediction. There's also talk about how Silver's performance is challenging to established punditry, as summed up in this XKCD comic.

Coming at this from the other side, though, I'm curious as a data person about how much secret sauce Silver's got. Sure, in broad qualitative strokes, he got the map right. But quantitatively, Silver's model also produced more detailed estimates about voting shares by state. How accurate were those?

Well, to start out, there is not some absolute sense of accuracy. When it comes to predicting which states would go to which candidates, it's easy to say Silver's predictions were maximally accurate. But what's tricker is to figure out how many he could have gotten wrong and still have us call his prediction accurate. For example, Ohio was a really close race. If Ohio had actually gone to Romney, but all of Silver's other predictions were right, could we call that a pretty accurate prediction? Maybe. But now let's say that he got all of conventional battle ground states right, but out of nowhere, California went for Romney. It's the same situation of getting one state wrong, but in this case it's big state, and an anomalous outcome that Silver's model would have missed. Would his prediction be inaccurate in that case? What if it was Rhode Island instead? That would be equally anomalous, but would have a smaller impact on the final election result. Now let's imagine a different United States where all of the races in all of the states had razor thin margins, and Silver correctly predicted 30 out of 50. In that case, we might say it was an accurate prediction.

All of this is to say that the notion of "accuracy" is really dependent upon what you're comparing the prediction to, and what the goal of the prediction is.

So what I want to know is how much Silver's model improves his prediction over what's just immediately obvious from the available data. That is, I want to see how much closer Silver's prediction of the vote share in different states was than some other baseline prediction. For the baseline, I'll take the average of the most recent polls from that state, as handily provided by Nate Silver on the 538 site. I also need to compare both the averaging method and the 538 method to the actual outcomes, which I've copy-pasted from the NPR big board. (Note: I think they might still be updating the results there, so I might have to update this post at some future date with the final tally.)

First I'll look at the Root Mean Square Error for the simple average-of-polls prediction and the 538 prediction. I'll take Obama and Romney separately. The "Silver advantage" row is just the poll averaging prediction divided by the 538 prediction.

	Obama	Romney
Averaging Polls	3.3	4.1
538	1.8	1.7
Silver Advantage	1.8	2.4

So it looks like Silver has definitely got some secret sauce, effectively halving the RMSE of the stupid poll averaging prediction. I also tried out a version of the RSME weighted by the electoral votes of each state, for a more results oriented view of the accuracy. I just replaced the mean of the squared error by a weighted average of the squared error, weighted by the electoral votes of the state. The results come out basically the same.

	Obama	Romney
Averaging Polls	3.2	3.1
538	1.5	1.5
Silver Advantage	2.2	2.0

So what was it about the 538 forecast that made it so much better than simply averaging polls? I think these plots might help answer that. They both plot the error in the 538 forecast against the error in poll averaging.

It looks like for both Obama and Romney, the 538 forecast did more to boost up the prediction in places where they outperformed their polls than tamping them down where they underperformed. The effect is especially striking for Romney.

So, Silver's model definitely outperforms simple poll watching & averaging. Which is good for him, because it means he's actually doing something to earn his keep.

You can grab the data I and R code was working with at this github repository. There's also this version of the R code on RPubs.

Teens and Texting and Grammar

noreply@blogger.com (JoFrhwld) — Fri, 27 Jul 2012 18:34:00 +0000

I'm just one man, one linguist, impotently shouting into the vast mediascape, "PLEASE POPULAR MEDIA! PLEASE DON'T RUN WITH THE TEEN TEXTING GRAMMAR STORY!"

There is a paper out in New Media and Society called Texting, techspeak, and tweens: The relationship between text messaging and English grammar skills. If you are a linguist, and you winced at the title, I have to warn you, you're not done wincing yet.

Is the key problem that the authors collected data on text messaging behaviors from self reports? No.

Is the key problem that the authors did not directly assess whether or not the teens in the study used "techspeak"? No. (Let's set aside the fact that high volumes of txtspeak are increasingly associated with out of touch adults).

Is the key problem that the authors didn't include any figures plotting the relationship between any of their measures? No.

Is the key problem that the authors included no control group of teens who don't text, or adults who adopted texting late in life? No.

The key problem is that the authors appear to have no idea what grammar or language are. I quote:

Similar to synchronous online communications such as instant messaging, the speed, ease, and brevity of text messaging have created a perfect platform for adapting the English language to better suit attributes of the technology. This has led to an evolution in grammar, the basis of which we shall call ‘techspeak.’ This language differs from English in that it takes normal English words and modifies them [...]

The depth of misunderstanding and naiveté present in this quote about the relationship between actual language and grammar and the way we write is equivalent to thinking that the sun revolves around the Earth, and that stars are bright dots on a large dome in the sky. Mind you, the Earth-centric, skydome model of the universe is a perfectly reasonable one until you are exposed to the most basic, rudimentary scientific understanding of how the world works.

The authors of this paper appear not to have been exposed to the most basic, rudimentary scientific understanding about how language and grammar work.

From Appendix A of the paper, I present to you the 20 point "grammar" assessment used in the study.

There (is, are) two ways to make enemies.
One of the men forgot to bring (his, their) tools.
Gail and Sue (make, makes) friends easily.
The coach thought he had (tore, teared, torn) a ligament.
During the flood, we (dranked, drank, drunk, drunked) bottled water.
The boy called for help, and I (swum, have swam, swam) out to him.
Fortunately, Jim’s name was (accepted, excepted) from the roster of those who would have to clean bathrooms because he was supposed to go downtown to (accept, except) a reward for the German Club.
I don’t know how I could (lose, loose) such a big dress. It is so large that it is (lose, loose) on me when I wear it!
The man around the corner from the sandlots (come, comes) to our meetings.
The man and his little girls (was, were) not injured in the accident.
The pictures in this new magazine (shows, show) the rugged beauty of the West.
The orders from that company (is, are) on your desk there.
The (boys, boys’, boy’s, boys’s) hats were lost in the water because they were careless in not tying them to the side of the boat.
(Its, It’s, Its’) an honor to accept the awards certificates and medals presented to the club.
Worried, and frayed, the old man paced the floor waiting for his daughter. (Correct/Incorrect)
The boy yelled, ‘Please help me’! (Correct/Incorrect)
She got out of the car, waved hello, and walked into the house. (Correct/Incorrect)
When Suzie arrived at the dance, no one else was there. (Correct/Incorrect)
Dad and I enjoyed our trip to new york city. (Correct/Incorrect)
The boy’s mother picked him up from school. (Correct/Incorrect)

To quote what it was the authors were trying to assess:

The first portion of the assessment consisted of 16 questions designed to test the student’s grasp of verb/noun agreement, use of correct tense, homophones, possessives, and apostrophes. [...] The second portion of the assessment asked participants to indicate whether or not a sentence was correct, such as ‘The boy yelled, “Please help me”!’ (Correct/ Incorrect). This portion tested the student’s understanding of comma usage, punctuation, and capitalization.

Virtually none of these points (homophones, apostrophes, comma usage, punctuation and capitalization) fall under the purview of what is scientifically understood to be "grammar". Arnold Zwicky has suggested the term "garmmra" for such things. Punctuation, comma rules, spelling conventions, etc. are all only arbitrary decisions settled upon a long time ago, and have nothing, nothing to do with human language. You could, by fiat, swap periods and commas (like many cultures do with their numeral systems), insist that sentence initial adverbs be followed by a semicolon, and decide to revert back to the symbols <þ> and <ð> to spell the sounds we currently both spell with <th>, and you know how many things that would change about English grammar? Zero things.

The remaining points of assessment could be considered to be well within the domain of grammar (tense and subject/verb agreement), except authors chose really poor, very variable items for the evaluation. The very first item involves verbal agreement with an expletive subject, and the rest involve cases of coordination, and agreement attraction! These are items which really lie on the outside edges of linguistic processing abilities, and there is no way that they could serve as reliable measures of fluency and grammatical competence. Search the work of any good writer, and I'm sure you'll find examples of both kinds of usage.

And then there's the second item: "One of the men forgot to bring (his, their) tools." Both possibilites are acceptable English, and have been for a long time.

The most depressing thing about this grammar assessment is where the researchers say they got it.

This assessment was adapted from a ninth-grade grammar review test.

I'm reminded of a piece I read called For Ebonics, the New Milennium Is Pretty Much Like the Old One, which said: "This suggests to me a catastrophic failure of the public school 'language arts' curriculum: people spend years in various language arts classes and leave with the same 19th-century folk notions that they started with."

So what have these authors actually found? Well, maybe it's the case that the more people who write in a broader range of contexts for a broader range of purposes, the more the arbitrary, conventionalized aspects of the writing system of English will undergo natural drift. What effect with this have on English grammar, as it is represented in the minds of every day English users? Probably just as much as the current writing system does: a minimal one.

And what about my plea to the popular media? Even if someone of note finds this post and reads it, I already know that it won't matter at all. Per my commentary on the coverage on vocal fry, no one is going to report on this piece because they care about science or facts. This research fits snugly into pre-existing biases about young people and the general decline of society, and frankly, these biases seem to have more to do with why these researchers did the study in the first place than science or facts. And there's is no way that something so trivial as a bunch of experts on language and grammar are about to derail this train of garbage and nonsense.

UPDATE! There is, in fact, actual paper on the topic of Instant Messaging and Grammar by Sali Tagliamonte and Derek Denis from 2008 called "Linguistic Ruin? LOL! Instant Messaging and Teen Language." Remember hearing about that in the news? Here's selections from their conclusions.

In a million and a half words of IM discourse among 71 teenagers, the use of short forms, abbreviations, and emotional language is inﬁnitesimally small, less than 3% of the data.

Our foray into the IM environment through quantitative sociolinguistic analysis, encompassing four areas of grammar and over 20,000 individual examples, reveals that IM is ﬁrmly rooted in the model of the extant language,reﬂecting the same structured heterogeneity (variation) and the same dynamic, ongoing processes of linguistic change that are currently under way in the speech community in which the teenagers live.

UPDATE! See also Enregistering internet language by Lauren Squires (2010)

Don't worry, I'm a physicist.

noreply@blogger.com (JoFrhwld) — Wed, 25 Jul 2012 19:11:00 +0000

Today, I came across a science news item from ABC (the Australian Broadcasting Corporation) with the title "Study opens book on English evolution." Oh goodness. Here are the opening paragraphs:

A study of 500 years of the English language has confirmed that 'the', 'of' and 'and' are the most frequently printed words in the modern era.

The study, by Slovenian physicist Matjaz Perc, also found the top dozen phrases most-printed in books include "at the end of the", "as a result of the" or "on the part of the".

That sound you hear is the stunned silence of linguists everywhere over the fact that you can get into the science news with the primary result that "'the' is the most common English word."

But to be fair, what the author was trying to argue is that the Zipfian distribution of word frequencies is a result of "preferential attachment," where frequent words get more frequent. He tried to demonstrate this by showing that the frequency of a word in a given year is predictive of its frequency in the future, specifically that relatively high frequency words will be even more frequent in the future. They key result is shown in Figure 4 in the paper, available here.

Say what?

While that quantitative result may stand, the fact that Perc is a physicist probably contributed to some really bananas statements about language. In the first paragraph, he almost completely conflates human language and written langauge as being the same thing, and erases the validity and richness of cultures with unwritten languages.

Were it not for books, periodicals and other publications, we would hardly be able to continuously elaborate over what is handed over by previous generations, and, consequently, the diversity and efficiency of our products would be much lower than it is today. Indeed, it seems like the importance of the written word for where we stand today as a species cannot be overstated.

He also presents some results of English "coming of age" and reaching "greater maturity" around 1800 AD (Figure 3). Finally! It only took us like, what, a thousand years or so?

The discussion section kicks off with the statement

The question ‘Which are the most common words and phrases of the English language?’ alone has a certain appeal [...]

That may be true for physicists, but for people who are dedicated to studying language (what are they called again?) not so much. Fortunately, his ignorance of linguistics is actually a positive quality of this research!

On the other hand, writing about the evolution of a language without considering grammar or syntax, or even without being sure that all the considered words and phrases actually have a meaning, may appear prohibitive to many outside the physics community. Yet, it is precisely this detachment from detail and the sheer scale of the analysis that enables the observation of universal laws that govern the large-scale organization of the written word.

See, linguists are just too caught up in the details to see the big picture! Fire a linguist and your productivity goes up, amirite?

For real though?

But back to the substantive claim of the paper. Is the Zipfian distribution of words due to the rich getting richer? That is, are words like snowballs rolling down a hill? The larger they are, the more additional snow the pick up, the even larger they get. Maybe, but maybe not.

Here's a little experiment that I was told about by Charles Yang, who read about it in a paper by Chomsky that I don't know the reference to. Right now, we're defining "words" as being all the characters between white spaces. But what if we redefined "words" as being all the characters between some other kind of delimiter? The example Charles used was "e". If we treat the character "e" as being the delimiter between words, and we apply this a large corpus, we'll get back "words" like " " and " th" and less frequently "d and was not paralyz". What kind of distribution to these kinds of "words" have?

Well, I coded up this experiment (available here: https://github.com/JoFrhwld/zipf_by_vowels) where I compare the ordinary segmentation of the Brown corpus into words by using white spaces to segmentations using "a", "e", "i", "o" and "u." Here's the resulting log-log plot of the frequencies and ranks of the segmentations.

It all looks quite Zipfian. So are not only the characters between spaces, but the characters between any arbitrary delimiters subject to a rich-get-richer process? Keep in mind that the definition of "word" as being characters between spaces is relatable to representations in human cognition, the definition of "word" as characters between arbitrary delimiters is not, especially not with English's occasionally idiosyncratic orthography.

Maybe it's possible for the results of my little experiment to be parasitic on a larger rich-get-richer process operating over normal words, but for now I'm dubious.