As part of Innovative Learning Week, we’ll be showing the award-winning adaptation of Jean-Dominique Bauby’s memoirs. Paralysed by a stroke, Bauby transcribed his memoirs letter-by-letter from his hospital bed, blinking as his interlocutor reached the right letter in the alphabet.
Both a powerfully emotional film and an insight into alternative communication methods employed by therapists working with severely speech impaired patients.
The Ghost in the Machine: Linguistics and Computation
Abstract: Almost sixty years ago, Chomsky 1957 placed automata-theory at the
centre of linguistic theory, arguing that natural languages were
beyond the context-free recognition capabilities of pushdown automata,
and raising the question of exactly what level of automata theoretic
expressive complexity would be the minimum needed to capture natural
languages. The interest of the question lies in the fact that most
natural linguistic phenomena, despite the important exceptions, seem to be
context-free, prompting the conjecture that there might exist a
“mildly context sensitive” natural family of languages with a little more
expressive power, but with comparably attractive computational
properties, and consequent increased linguistic explanatory adequacy.
Transformational rules themselves turned out to be too expressive to
be of automata theoretic interest in this sense, and mainstream
linguistics has shown little interest in the question since. However,
in recent years there have been a number of proposals for less
expressive formalisms from computational linguists who build practical
devices for tasks like question-answering and machine translation.
I’ll review some of these developments in non-technical linguistic
terms, using examples from various languages, and draw some
conclusions for understanding problematic notions such as universal
grammar, the role of statistical models, and the course of language
acquisition in children as observed by psychologists.
The talk will start at 6:00 p.m. and last about 1 hour. It will be followed by a Q&A session (about half an hour). We will then go to a pub for food and drink with the speaker.
Our talks are public lectures open to all, regardless of whether you are a student or not or what or where you study if a student. We aim for all of our events to be accessible to all; please feel free to contact us beforehand if you require assistance or further information.
- See more at: http://langsoc.eusa.ed.ac.uk/?cat=9#sthash.cp7rOBq1.dpuf
Langsoc hasn’t done a reading group for a while, so we thought we’d do this as something a bit different.
Just have a look at the paper and come along for a relaxed discussion over a drink at the Potting Shed pub. You can contribute as much as little as you like, and don’t worry if you don’t have time to read the whole paper—you can just skim read it: all will be explained at the event. The discussion will be led by our Vice-President, who has a strong interest in the linguistics of language revival and personal experience of the subject.
Some days ago I had a sensation in my eyes which I referred to as ziepen (tweaking) when I told a (German) friend. She got what I meant even though she didn’t actually know the word and it would not normally be used in this way. I wondered:
How did I come up with ‘ziepen’ if it wasn’t meant to be used in such a context?
Ziepen describes ‘the brief pain through pulling of hair’. The meaning of ‘brief pulling pain’ can probably be transferred easily, e.g. in my situation to my eye.
How did she know what the word meant for me, i.e. what kind of sensation I was trying to describe, without actually knowing it?
New words are often deduced from words with similar meanings, giving them a similar sound pattern, which belongs to the study of phonesthemic patterns. A phonestheme is a sound (sequence) that suggests a certain meaning and has the tendency to show up in neologisms. Phonesthemes are often initial, but can also be final or even medial. The remainder of the word may not itself be a morpheme. (Waugh, 1994) Some examples:
sn- (found in nose-related words): sniff, snore, snot, sneer, sneeze…
The German cluster zw(cf. tw in English) is generally related to the number two: zwei (two), zwischen (between), Zwilling (twin)…However, there are several exceptions, e.g. zwirbeln (twirl) and zucken (twitch). Although these don’t carry the specific meaning (two), they do seem to be somewhat semantically related.
Studies have shown that when asked to invent/interpret new words, participants look at phonesthemes in their language to follow a predictive pattern. This could explain why my friend was able to tell what I meant without actually knowing the word: She might have associated ziepen with zucken which describes a brief, usually involuntary movement.
In general, how do we agree on words for particular sensations while we can’t tell what they feel like for anyone else, and how do we learn them?
When we learn new words we understand their meanings by, for example, being pointed at the ‘thing’ or the ‘action’ they (nouns and verbs) describe. When it comes to internal experiences like perception it all gets a lot fussier.
How do you know that the pain I experience is actually painful? This touches upon linguistics, philosophy and biology. All our knowledge comes in through our senses which are subjective and unreliable (as revealed by the many ways in which they can be manipulated and deceived, e.g. hallucinations and optical illusions) and truth is defined by language which is more of an agreement than an objective state.
Pears (1971) explains that an empiricist view on sensations involves accepting that the general meaning of e.g. ‘pain’ involves two aspects: 1) the set of teaching links 2) the inner reference (private sensation). Only a primitive empiricist would think it only involves the latter – a language only about private sensation would be unteachable. He says that “our language of sensations is not really teachable, and we do not ever really communicate about such matters” (p.158). So, while we can ‘teach’ nouns and verbs that can be pointed at in the world, we cannot teach words that have their meaning lying within ourselves. We all live in our own worlds, we can never know for sure about other people’s sensations, so I guess instead of being ‘taught’ we can only infer such information from cues: When I bleed, I’m hurt and I’m in pain. So when you bleed, you are probably experiencing pain too.
Yet it is to note that pain comes in a whole lot of different forms, as you may notice when you see your doctor and try to find words to explain what’s hurting you. That’s where neologisms and phonesthemic patterns might come handy…
It is incredible how we manage to communicate things we don’t have words for, by using common words or inventing new ones. Life is miraculously mysterious and mysteriously miraculous!
Oxford handbook of Wittgenstein (2011). Oxford: Oxford University Press.
Pears, D. (1971). Wittgenstein. London: Fontana.
Waugh, L. (1994). Degrees of Iconicity in the Lexicon. Journal of Pragmatics, 22(1), 55-70.
Correctly speaking, Gin ‘n Tonic. While most people probably know that the ‘n stands for ‘and’, eggcornishly it could also stand for ‘in’, after all, the gin end up in the tonic water, right!?
Now, when I heard the drink’s name for the first time, I thought it was ‘Gin a’ tonic’. Well, connected speech and perception (and maybe even ‘nativisation’ in my case, i.e. making it sound more like ‘the perfect drink for me’ (note my name)) can do wonderful things to word sequences, in this case even clarify that the connective is and, and not either and or in…
(Urban Dictionary for ‘ginatonic’: This state of being is induced by drinking excessive amounts of alcohol, and becoming non-responsive to any external stimulus…)
What would we do without those little symbols that organise our words on paper, that indicate intonation and thus meaning, essential to disambiguate sentences?
Earliest alphabetic writings had no capitalisation, no spaces, no vowels (i.e. today’s Abjad writing system) and very few punctuation. In ancient Greece and Rome, symbols were used to mark where speakers should pause; the longest section (the period) was divided by the ‘period’ pause – hence a ‘period’ mark –, the shortest section (“that which is cut off”) was divided by a ‘comma’.
When the Christian Bible started to be copied and translated (around 400AD), an early punctuation system was developed. Only when printing was introduced in the 15th century however, punctuation became more common and systematic.
Nowadays, punctuation varies across languages, register, and authors may (not) use it for a specific stylistic style.
While punctuation may facilitate comprehension, the lack of it may facilitate ambiguity – which must not be a bad thing at all.
Imagine you had a fight with your boy/girlfriend and you’re mad at them but don’t want to show it; you leave them a note saying ‘Have a good day’ without any punctuation. Many ways to read it:
Have a good day! – you fool
Have a good day… with me? Surprise me with some chocolate and apologise
Have a good day. There’s nothing more to say, just give me time to breathe.
Well, something along those lines. Makes the whole story much more interesting, don’t you think!?
Dale, R. 1991. The role of punctuation in discourse structure. In Proceedings of the AAAI Fall Symposium on Discourse Structure in Natural Language Understanding and Generation, 13 14. USA.
Lee, S. 1995. A syntax and semantics for text grammar. MPhil. Dissertation, Engineering Dept., Cambridge University.
On one of our first days here at the University of Auckland (NZ) we had a Maori welcome: a haka (war dance) and a Maori greeting us in Maori. It more or less sounded like “hakamuto…”, a string of just a few sounds. Indeed, Maori, an Eastern Polynesian language, has only 10 consonants: h, k, m, n, p, r (rolled), t, w, ng (velar nasal), wh (pronounced as f), and 5 vowels (vowel length is phonemic): a, e, i, o. u.
At the end of the 19th century, Maori became a minority language. Fewer people spoke and learnt to speak the language. Only at the end of the 20th century, the dangers of the language loss were recognised and recovery programmes were initiated, and Maori became New Zealand’s second official language in 1987. After a brief revival, however, the language has seen another decline in speakers – only about 9% of the Maori population is fluent in the language (about 4% of the population).
Maori words have their place within Kiwi-English. Many place names have not been translated into English and tell of the landscape’s properties, for example ‘Aotearoa’ means ‘cloud white long’ (land of the long white cloud, the Maori name for New Zealand). More examples can be found here.
Kiwi-English is similar to Australian English. What I notice most is the shift from ‘e’ to ‘i’. Further differences are that the short ‘i’ has centralised towards schwa, and the short ‘a’ sound has moved towards the short ‘e’. I was a bit scared I wouldn’t understand a thing, but I haven’t had any troubles yet.
So much for the linguistic side of down down under. I’ve been here for a week now, lectures start tomorrow, so there’ll be more to tell soon.
(-wait for it-) If you haven’t heard of How I met your mother yet… you’re missing out. I’m not an addict of the series (as some of my friends) but I really enjoy watching it, it’s fun, a wide range of people can relate to the different characters and it even deals with more serious issues such as racism. But the main reason for a linguist to watch it is… the language, of course. Just take this scene – wouldn’t it be awesome if people came up with such an amazing wordplay in everyday life, rather than reading it off a script!?
Jerry: So, Saturday night? Time to cut loose, right? Who wants to split a beer? Barney: Oh, we’re not drinking here. Tonight, we’re going big. Let’s see, what club should we hit first? There’s club Was, there’s Wrong… Marshall: Um, those places shut down a long time ago. Barney: Oh no… Marshall: Oh No shut down too. Ted: There’s Where. Jerry: Where’s Where? Lily: Where’s where Was was, isn’t it? Barney: No, Was wasn’t where Where was, Was was where Wrong was, right? Jerry: Ok… Ted: Not OK, that place is lame. Robin: OK is Lame? I thought Lame was a gay bar… or is that wrong? Marshall: That’s Wrong. That’s not wrong. Barney: Guys, focus. Robin: Oh, I like Focus! Let’s go there. Ted: Where? Robin: Not Where. Focus! Lily: I thought Focus was closed. Barney: No, Was was Closed. Once Was shut down, it re-opened as Closed. Marshall: So Closed is open. Robin: No, Closed is closed. …
Apart from scenes like that, the series has brought quotes into some of my friends’ lives that they could no longer live without (mostly Barney-phrases)…
“organ of hearing,” O.E. eare “ear,” from P.Gmc. *auzon (cf. O.N. eyra, Dan. øre, O.Fris. are, O.S. ore, M.Du. ore, Du. oor, O.H.G. ora, Ger. Ohr, Goth. auso), from PIE *ous- with a sense of “perception” (cf. Gk. aus, L. auris, Lith. ausis, O.C.S. ucho, O.Ir. au “ear,” Avestan usi “the two ears”). The belief that itching or burning ears means someone is talking about you is mentioned in Pliny’s “Natural History” (77 C.E.). Until at least the 1880s, even some medical men still believed piercing the ear lobes improved one’s eyesight. Meaning “handle of a pitcher” is mid-15c. (but cf. O.E. earde “having a handle”). To be wet behind the ears “naive” is implied from 1914. Phrase walls have ears attested from 1610s. Ear-bash (v.) is Australian slang (1944) for “to talk inordinately” (to someone).
O.E. heran (Anglian), (ge)hieran, hyran (W.Saxon) “to hear, listen (to), obey, follow; accede to, grant; judge,” from P.Gmc. *hausjan (cf. O.N. heyra, O.Fris. hora, Du.horen, Ger. hören, Goth. hausjan), perhaps from PIE *kous- “to hear” (see acoustic). For spelling, see see head (n.); spelling distinction between hear and heredeveloped 1200-1550. O.E. also had the excellent adjective hiersum “ready to hear, obedient,” lit. “hear-some” with suffix from handsome, etc. Hear, hear! (1680s) was originally imperative, used as an exclamation to call attention to a speaker’s words; now a general cheer of approval. Originally it was hear him!
Taken from the etymonline.com dictionary. Fun times, eh?
m tryng t prf pnt hr tht ppl wh knw nglsh r prfctl (r s) cpbl f ndrstndng txt wrttn wtht ny vwls. hv lrdy wrttn pst rltng t ths r n rthgrphy (hr). Snc w’r t t tryng t ‘mprv’ nglsh rthgrphy, whch s n f th lst phnmc ns, t s rgbl whthr rthgrph nd phnlgy shld b cmbnd t t lst sm xtnt n ths pst s .g. y s phnlgclly vwl, bt n th bc t s rgrdd s cnsnnt. D t smplcty, wll nt dlt t hr thgh.
Tr, t s qt hrd t ndrstnd shrt wrds tht, thrgh vwl dltn, nd p s sngl cnsnnts nd nly th cntxt mks clr wht s mnt (bt t my stll rmn mbgs), nd sm wrds lk th ndfnt rtcl r cmpltly rdctd. Hwvr lnggs sch s rbc shw tht t s pssbl – nd nglsh lrdy mks s f t n fr xmpl txts.
Wld vwl dltn llw rthgrph t rprsnt nglsh nd ts dlcts bttr? ftr ll, th dffrnc n th prnnctn f vwls s mjr chrctrstc f dlcts nd ccnts. Frthrmr, ths sstm wld shrtn txts lt. Nvrthlss, t lst whn y’r nt sd t t, t tks mch lngr t wrt nd rd.
wll kp ths pst shrt (s t s tkng m whl t tp nd prbbl y w bt t fgr t). hp cld 1) mk y thnk bt nglsh rthgrphy (phnlgy nd th PA), 2) shw hw mzng th hmn brn s n bng bl f dcphrng (hpflly mst f) ths, nd 3) gv y lttl prcrstntn brk frm wrk
(shd hv smply wrttn p ths pst ‘wth vwls’ nd thn dltd thm, nd nt wrttn t strght wtht vwls (=tkng lngr nd pssbly csng (mr) mstks), bt t ws gd brn xrcs!)
I’m trying to proof a point here that people who know English are perfectly (or so) capable of understanding text written without any vowels. I have already written a post relating to this area in orthography (here). Since we’re at it trying to ‘improve’ English orthography, which is one of the least phonemic ones, it is arguable whether orthography and phonology should be combined to at least some extent in this post as e.g. y is phonologically a vowel, but in the Abc it is regarded as consonant. Due to simplicity, I will not delete it here though.
True, it is quite hard to understand short words that, through vowel deletion, end up as single consonants and only the context makes clear what is meant (but it may still remain ambiguous), and some words like the indefinite article are completely eradicated. However languages such as Arabic show that it is possible – and English already makes use of it in for example textese.
Would vowel deletion allow orthography to represent English and its dialects better? After all, the difference in the pronunciation of vowels is a major characteristic of dialects and accents. Furthermore, this system would shorten texts a lot. Nevertheless, at least when you’re not used to it, it takes much longer to write and read.
I will keep this post short (as it is taking me a while to type and probably you a wee bit to figure out). I hope I could 1) make you think about English orthography (phonology and the IPA), 2) show how amazing the human brain is in being able of deciphering (hopefully most of) this, and 3) give you a little procrastination break from work
(I should have simply written up this post ‘with vowels’ and then deleted them, and not written it straight without vowels (=taking longer and possibly causing (more) mistakes), but it was a good brain exercise!)
Edinburgh University Linguistics & English Language Society