Linguists on the job market
« previous post | next post »
Chris's post on the job market for linguists circulated instantly here at the University of Arizona, and one of our enterprising recent grads, Shannon Bischoff, thought of comparing Chris's job posting numbers, sorted by area, to numbers of dissertations produced in each area. I post his revised table of figures including the diss numbers below the jump.
Shannon writes:
"…below each of the job numbers [in the right-hand column, from Chris's table of job ads from 2008] is the number of dissertations listed at proquest in the last 5 years in [each] area…note proquest lists 3124 dissertations in linguistics in the last 5 years in total (I believe I controlled for dissertations in English, and used the subfields as search terms with "linguistics")."
And here's the result of his count:
| Job Ads: Dissertations | |
|---|---|
| Computational linguistics | 34 (# of job ads targeting this specific area in '08, from Chris's post)
110 (dissertations in last 5yrs) |
| Applied linguistics | 20
38 |
| Syntax | 12
749 |
| Phonology | 9
290 |
| Semantics | 6
262 |
| Psycholinguistics | 4
56 |
| Sociolinguistics | 4
152 |
| Phonetics | 4
79 |
| Cognitive science | 4
20 |
| General linguistics | 22
(no dissertations in 'general linguistics') |
Shannon cautions me to warn that his survey of numbers on proquest was not 'scientific', so consider yourselves duly warned! But there are some interesting discrepancies in the ratios of jobs available to ratios of dissertations produced; compare syntax, e.g, with job:diss ratio of 12:749 (0.016), to applied linguistics, with a job:diss ratio of 20:38 (0.526).
And let me just echo Chris's amazement that every 4-year college in the US doesn't have a linguist on the faculty. Linguistics is tailor-made for courses that provide a stimulating, analytical, scientific research experience in the absence of significant amounts of expensive lab equipment or external funding. Students are excited about language and usually surprised and intrigued to discover its hidden complexities as well as to learn more about the more obviously attention-grabbing topics… and language is one of the core human capacities‚ surely suggesting it should be included on the curriculum of any humanistically-oriented undergraduate degree program. Four-year colleges, you're missing a golden opportunity!
Allison said,
January 26, 2009 @ 4:17 pm
So, wait, does this mean I *should* get a post graduate degree in linguistics or that I shouldn't?
Am I going to be able to get a job when I'm done?
marie-lucie said,
January 26, 2009 @ 4:23 pm
I wonder how the ratio of jobs vs graduates in linguistics compare with those in other (non-business) disciplines?
Miriam Meyerhoff said,
January 26, 2009 @ 4:24 pm
Chris' original posting was very interesting and his figures — averaged over ten years could actually be useful. However, I think the figures Heidi/Shannon post are likely to be less useful since they refer only to one year. Linguistics is not a huge field (as the comments bemoaning the fact that it is not represented in every college attest), and as we all know statistically non-significant fluctuations in small field can look extreme. I would be pretty dismayed if someone was planning their area of specialisation, or if a department was planning their next hire, on the basis of the dissertations:jobs ratio for 2008 only that are shown above.
Heidi Harley said,
January 26, 2009 @ 4:45 pm
Oh! I made a mistake then, Miriam — the job figures are the exact ones given for 'ads targeting specifically one area' from Chris's post, so if those figures are averages over 10 years, then that's what they are, not just the '08 figures. I'll edit the table above to take out reference to '08. Sorry about that!
kyle gorman said,
January 26, 2009 @ 4:48 pm
this would really benefit from graphical presentation. i suggest a plot of either the ratios, or a scatter plot with a linear regression line, then looking at the outliers. i'd do it but i have to go to a meeting
marie-lucie said,
January 26, 2009 @ 4:52 pm
I notice that the various specialties do not include historical linguistics, even though this is the part that non-linguists tend to be especially interested in.
Nathan Myers said,
January 26, 2009 @ 5:04 pm
Allison: You should get a post graduate degree in linguistics, and then persuade a college that disgracefully lacks any semblance of a linguistics department to hire you to start one. In fact, you should at that time arrange for several colleges to enter into a bidding war over which may avail itself of your services in that capacity. As there are evidently many such colleges, the competition should be fierce.
Heidi Harley said,
January 26, 2009 @ 5:26 pm
Also, just to be clear — the numbers look scarier than they are: The diss #s are for dissertations over a 5 year period, and the job #s are for average # of jobs in that area over a *one*-year period, so the ratio is not representative of the market in a given year! Divide the diss #s by five to get a truer picture of the competition, maybe.
Chris Potts said,
January 26, 2009 @ 5:27 pm
Miriam Meyerhoff's comment is definitely worth keeping in mind; it is easy to imagine misuses of the numbers. However, those numbers, together with Heidi and Shannon's, paint an accurate (partial) picture of the current state of the market. People within five years of earning their PhDs are likely to be still checking the ads, looking for work, considering a move, etc.
Chris Potts said,
January 26, 2009 @ 5:28 pm
marie-lucie!
The full list we are working with is here; historical linguistics was below the threshold I set for display. (The original post links to that file too.)
marie-lucie said,
January 26, 2009 @ 5:30 pm
Linguists in academia are not all in linguistics departments, as those are relatively few. Many people with either PhD's or substantial training in linguistics are teaching in departments of English or other languages. It helps a linguist to have expertise to teach in one of those areas (or others, eg psychology, anthropology, sociology) and to introduce linguistics courses that way. This is a more realistic plan than trying to create a department when none existed before.
Ran Ari-Gur said,
January 26, 2009 @ 8:43 pm
@marie-lucie: Indeed, my alma mater has no Linguistics program, but its Department of Cognitive Science has enough linguists on its faculty that one could be excused for thinking "cognitive science" was a euphemism for "linguistics"!
(I don't know if the links in this comment will survive the filter; if not, you can always Google my name to see what my alma mater is.)
marie-lucie said,
January 26, 2009 @ 9:24 pm
Ran, thank you for the links, I see that there is an MA in cognitive linguistics, and that the fundamental linguistics course is listed under English. I think that there are many more linguistics courses around even small institutions than just the ones listed as LING, so recent graduates should not necessarily limit their job search to Linguistics departments but read the catalogues carefully.
Christopher Stone said,
January 26, 2009 @ 9:45 pm
My own experience mirrors Marie-Lucie's; I'm an undergraduate at Truman State University, which has had linguistics courses (and majors) for a long time now but until recently has been housed in the English department (now called the Department of English & Linguistics).
pc said,
January 26, 2009 @ 10:27 pm
Ack Ack Double-Ack, etc. The issue of linguistics being such a small field has always been both exciting and troublesome to me, knowing that many of the "other" (meaning, non-first-tier/R1/tenure-track) academic options (teaching colleges, community colleges, small universities, etc.) that are available to more "core" humanities or social science phd's are kind of by default not as open to me.
But I wonder, how quickly could a field grow? I mean, the implementation of a newly established/titled Linguistics department (or co-department) like the one mentioned for Truman–how often does that actually happen? Can we get breakdowns of how many higher ed institutions have a) no linguistics representation, b) linguistics rep only in the form of one or two courses taught by one or two faculty members, c) linguistics rep only in the form of an interdepartmental program or core of faculty members, d) linguistics departments? I'm sure there's an easy source to consult for this, but I am far too lazy.
Anyway, as a mid-program phd candidate, I wonder what the real chances are for growth of the field in the next few years (esp. given the economic turn), and if there's anything students, faculty, or our professional associations can do (or are doing) to encourage adoption of more linguistics teaching? (I ask completely unironically.)
language hat said,
January 27, 2009 @ 10:12 am
Am I going to be able to get a job when I'm done?
I left the field thirty years ago in part because of the lack of jobs for new PhD's, and the situation doesn't seem to have improved much. (Other reasons: going into too much debt, not enjoying teaching, and hating my dissertation.)
Eric Baković said,
January 27, 2009 @ 10:44 am
Honestly curious here: are numbers of applicants for particular jobs a matter of public record (at least, at public institutions)? It would be good to contrast the numbers above with some numbers that show how many folks are actually competing for individual jobs.
dw said,
January 27, 2009 @ 12:11 pm
Apologies if this is off-topic, but one sentence of the post struck me:
And let me just echo Chris's amazement that every 4-year college in the US doesn't have a linguist on the faculty.
To me, this would indicate that no 4-year college in the US has a linguist on the faculty, rather than the presumably intended meaning that some (but not all) such colleges lack a linguist.
In order to communicate the intended meaning, I would say "not every 4-year college … has a linguist on the faculty" or, less awkwardly, "not all 4-year colleges … have a linguist on the faculty". I grew up in the UK: is this a UK vs. US thing?
Arnold Zwicky said,
January 27, 2009 @ 12:53 pm
To dw: the relative scoping of negation and a universal quantifier is a topic that comes up every so often on Language Log, usually when someone objects to sentences intended as having wide-scope negation (conveying 'it's not the case that every …', rather than 'for every …, it's not the case that …'). Mark Liberman took on such an objection from James Kilpatrick a while back, citing wide-scope negation in the works of various excellent writers (in the belief that his own judgment, and that of Neal Whitman, that the wide-scope negation reading for Kilpatrick's example — "Mass Transit Not an Option for All Drivers" — was the preferred reading would not satisfy Kilpatrick).
Actually, Mark noted that "In fact, after a modest amount of searching, I haven't come across a single published example where a competent writer of English follows Kilpatrick's theory of semantic interpretation. There must be some out there …"
(And no, it's not a US vs. UK thing.)
Heidi Harley said,
January 27, 2009 @ 3:24 pm
Re the appearance and disappearance and appearance of the "in '08" restrictor above—Chris wrote to say that his numbers concerning job ads were indeed just for '08, not averaged over 10 years. So Miriam's comment concerning validity of the numbers of jobs available over time still stands, but it applies (I guess) to the original post as well as to this followup!
dw said,
January 27, 2009 @ 6:18 pm
Arnold Zwicky:
I don't think that any of the examples in Mark Liberman's post is quite the same as the sentence here. They are all of a type exemplified by:
Mass transit [is] not an option for everyone
where the word "not" or some other negative precedes the quantifier. But
[e]very 4-year college in the US doesn't have a linguist on the faculty
is different. For me, the relative positions of the negative and the quantifier seem to be important. I don't have a problem with any of Mark's examples: it's only the sentence here that seems odd. I'm curious to see whether anyone else has the same reaction.
Claire Bowern said,
February 3, 2009 @ 1:07 pm
Re the syntax number, depending on how the counting was done it might be a bit inflated by people like me who listed 'syntax' as a PhD field even though my dissertation was on [comparative-historical Nyulyulan morpho]syntax; historical was the primary field and syntax a secondary one.
John McCarthy said,
June 9, 2009 @ 7:55 am
These data have gotten a lot of attention, but the counts of dissertations seem to be just plain wrong. I tried to reproduce the results, and the only way I came close is by using all of Proquest's defaults (except for the 5-year span) and entering "syntax", "semantics", etc. (minus the quotes) in the first field on the search page. Unfortunately, the defaults give you a search for instances of this word occurring anywhere in the abstract, in any field, in a dissertation or master's thesis. Only a modest fraction of the hits are to anything that a linguist would call a syntax or semantics dissertation. The rest include things like the syntax of a programming language or even a dissertation in musicology.
A better search strategy is to look for the Index term "syntax", "semantics" etc. in the Subject field Linguistics in dissertations only. Using this method, I get 5-year counts of:
Syntax: 180
Semantics: 113
Phonology: 110
There are still some errors, I'm sure, but at least we're not seeing the gross overcount in the original data.