bookmark_borderHans Arndt in memoriam

Jakob Steensig havde flg. triste notits i det seneste nyhedsbrev fra det fhv. Institut for Lingvistik:

Hans Arndt døde i onsdags på Djursland Hospice. Hans Arndt var lektor på lingvistik fra midten af 1980’erne til han holdt op for en del år siden. Hans var den som genstartede lingvistik efter at det næsten var forsvundet. Gennem en utrættelig indsats og en helt usædvanlig åben og dialogisk stil lykkedes det ham i løbet af ti til femten år at få lingvistikfaget, hvor der nogle år i firserne var to eller nul studerende som startede hvert år, og næsten kun ham der underviste, til at blive en succes, med nu seks faste undervisere og langt flere studerende som søger, end der er plads til.

Hans var lingvistik i lang tid. Og han var en utroligt vigtig inspiration for mig og mange andre. Jeg håber at jeg og andre der kendte ham, får lejlighed til at fortælle nye studerende og andre ved lingvistik mere om ham og hans inspiration, men lige nu ville jeg bare sikre mig at alle som kendte Hans, får de sørgelige nyheder.

Jakob har helt ret. Da jeg begyndte på lingvistik i 1990, var Hans Arndt reelt hele den lingvistiske del af instituttet. (Der var naturligvis også en ungarsk og en finsk lektor, da det jo dengang hed Institut for Lingvistik, Finsk og Ungarsk – eller Institut for Lingvistik og Underlige Sprog, som det blev kaldt i folkemunde.) På papiret var der en lingvistiklektor til, men han var udlånt til Jydsk Telefon, så i praksis skulle Hans trække hele læsset selv, kun hjulpet af undervisningsassistenter af svingende kvalitet. Det tog en del år, før bemandingen endelig kom op på et fornuftigt niveau, og det var fantastisk flot, at han klarede at holde det kørende i alle årene.

Hans Arndt var en særdeles dygtig underviser, og han var god til at få det bedste ud af en vanskelig situation – fx fik mange af de studerende mulighed for at være undervisningsassistenter i løbet af uddannelsen (jeg underviste fx i Lingvistiske Teorier før 1920 i forårssemesteret 1996), og det lærte vi meget af.

I sine sidste arbejdsår udvidede han sine undervisningsnoter i Sproglig Analyse til en glimrende populærvidenskabelig bog, Sproget: hverdagens mirakel, og det er en bog, som bør stå i ethvert hjem.

Hvad mindes jeg ellers? Hans fine russiskudtale (han var tidligere sprogofficer og havde i 1968 siddet klar i et jagerfly, da de sovjetiske tankvogne rullede ind i Prag). Lejrbål på plænen på hans nedlagte gård. Og hans fine sangstemme, som han gerne brugte til at synge Poul Henningsens sprogsang med:

For mig er sprogets klang
min mors stemme
kort og klart som hammerslag
med venligt sving.
Hver glose blank og rund
og go’ at ta’ på
alle skarpe kanter slidt
ved hverdagsbrug.

De grønne marker,
det krappe sund,
blå solskinstimer
og månens segl.
Alt det har osse ret.
Hver ting til sin tid.
ord skal være redskab først –
og så musik.

Hør jydens seje stød.
Hør øernes syngen
Hør Køvnenhavnerdrengens ‘a’
når har si’r Far.
Det’ livets tumleplads
det’ sprogets havstok.
Her blir tiden skuret til
og vasket ren.

Det fine menneske,
det sjældne digt
La’ dem beholde,
det skrevne ord.
Men det er brugsværdien
i din og min mund.
Sproget står og falder på:
Det talte ord.

Mindet om ham vil leve længe.

PS: Jeg tror desværre ikke, jeg har noget billede af Hans, som jeg kan bruge til at illustrere denne nekrolog med. Hvis du har et, jeg må bruge, må du meget gerne sende mig det!

bookmark_borderEmil fra Lønneberg og Julemanden

Emil og Julemanden.
Anna (som lige er fyldt ti) læser hver aften lidt op for mig for at blive bedre til dansk (og jeg læser også højt for hende). For tiden læser hun Emil fra Lønneberg, og det går da også ganske godt.

Nogle gange går det dog galt, som for eksempel, da hun glad og fro sagde flg.:

Emil spejdede op i skorstenen, og da så han noget sjovt. I hullet lige over hans hoved hang en rød julemand og kiggede ned til ham.

„Hej med dig,“ sagde Emil. „Nu skal du se en, der kan klatre!“

I originalen står der „julimåne“, men det er nu ikke nær så sjovt!

På samme måde læste hun flg. et par sider senere, men det var nu måske nok med vilje, for hun gjorde det med et skælmsk smil:

Men i Katholtsøen mellem hvide åkander svømmede Emil og Alfred rundt i det kølige vand, og på himlen hang julemanden, rød som en lygte og lyste for dem.

„Dig og mig, Alfred,“ sagde Emil.

„Ja, dig og mig, Emil,“ sagde Alfred, „Det skulle jeg mene!“

Phyllis bestemte sig i øvrigt for at teste Léon på den første passage, og han begik den selvsamme fejl som Anna, så det må være en oplagt fejl for dansk-skotter.

bookmark_borderScots on Smartphones

Writin Scots uisin SwiftKey’s preditive keyboard.
A’ve been fasht for a lang time at predictive keyboards wadna recognise Scots ava – ilka time ye uised a perfecklie normal wird, it wad get chynged tae a completelie different Inglis wird at juist happent tae leuk similar.

Sae A wis weel chuft whan ane o ma clients, Scottish Language Dictionaries, gat a email fae Julien Baley fae SwiftKey (a Lunnon-based companie awnt bi Microsoft) twa-three months syne anent addin Scots tae thair predictive keyboard for Android an iOS. A dae aw the data stuff for SLD, sae o coorse A wis chosen tae wirk wi Julien on this.

A extractit the relevant bits fae the new edition o the Concise Scots Dicionary an sent this tae Julien. Forby, A gied him a earlie version o a corpus (a collection o texts) o modren Scots. He separatelie contactit Andy Eagle and gat the heidwirds fae his Online Scots Dictionary.

Suin efter this, Julien sent me the first version o the keyboard. At this pynt, it daedna ken the Scots inflections, an it wis makkin some unco substitutions (e.g., aA oweraw), sae A advised him on the grammar o Scots an on the substitutions. The final bit wis tae leuk at wirds he fund in the corpus at wisna in the dictionars, an the keyboard wis redd.

Ye can doonlaid SwiftKey on yer Android smartphone the day, but gin ye hae a iPhone, ye maun wait few mair days (technical issues pat it aff).

SwiftKey will lair fae the wey fowks uise it, sae it’ll get better and better.

A howp this will see monie mair fowks writin Scots wi confidence, an ultimatelie tae better support for Scots in programs an on wabsteids. Wad it no be great gin Scots wis supportit in yer spellchecker, in Google Translate, and as Facebook interface leid?

PS: A wis chuft tae sae stories aboot this in The National, The Herald an Bella Caledonia.

bookmark_borderGeneration X are disappearing

Hugo Rifkind had an article in The Times today about being a Xennial (too old to be a Millennial, too young for Generation X), and I sent him the following tweet as a reply:

Hugo and 77 other people (so far) were kind enough to like it, so I thought I’d elaborate a bit on my theory.

A lot of the stuff about the Baby Boomers, Generation X and the Millennials can be traced back to Howe and Strauss’s Generations from 1991. This book examined earlier American generations and claimed to identify a four-generation cycle. They then defined the new generations that were emerging at the time and tried to predict their future very roughly. In particular, they expected a huge crisis once the Baby-Boomers had started to retire (perhaps around 2020), which Generation X would sort out and then hand over power to the Millennials.

This is clearly not what happened – the crises (9/11 + the financial crash) happened much sooner than they expected, while the Baby-Boomers were still in office. They actually mentioned this possibility briefly on page 382:

What happens if the crisis comes early? What if the Millennium – the year 2000 or soon thereafter – provides Boomers with the occasion to impose their “millennial” visions on the nation and world? The generation cycle suggests that the risk of cataclysm would be very high.

Furthermore, in their historical analysis they clearly don’t assign a standard length to generations, so they would themselves have expected the generational boundaries in the 20th century to require some tweaking once the big defining events had taken place. It’s therefore completely in their spirit to revisit the definitions they suggested more than 25 years ago.

They actually don’t even stick to four generations per cycle all the time. What they call the Civil War Cycle contains only three. As they write on page 192:

[It is] America’s only three-part cycle – the one whose crisis came too soon, too hard, and with too much ghastly devastation. This cycle is no aberration. Rather, it demonstrates how events can turn out badly – and, from a generational perspective, what happens when they do.

I’m postulating that this has happened again. The crisis came so soon that at least half of Generation X hadn’t yet managed to get high enough up the housing ladder (or build up assets in other ways) to allow them to benefit from the asset boom that was a result of the financial crash. As a result we now have a huge split in most western societies: On the one hand, older people (Baby Boomers and older X’ers) often are asset-rich and have paid off most of their house, as well as having a good pension. Other members of this generation are less rich, but they might at least have a cheap council house that is affordable on their salary or their pension. On the other hand, younger people (Millennials and younger X’ers) don’t tend to have much wealth: They’re either renting in the private sector, or they’ve paid so much money for their house that a crazy amount of their salary is spent on the mortgage. They don’t have decent pensions, and they don’t really expect ever to be able to retire comfortably. They also typically grew up being told to expect a great and prosperous life, and they weren’t expecting things to turn out like this.

I was born in 1972, so right in the middle of Generation X, and I think we felt different from both the Baby Boomers and the Millennials before the financial crash. However, I now feel more and more similar to the Millennials, and further and further removed from the Boomers. So I think we might have to redefine the Baby Boomer generation as stretching all the way to the late 1960s, and the Millennials starting immediately afterwards. (I don’t believe it’s a clean break – whether somebody belongs in one generation or the other ultimately depends on whether they had enough assets when the economy collapsed.)

I think we can now also tell when the Millennial generation ended: The youngsters who don’t remember the time before the financial crash have a different mindset because they didn’t spend their childhood expecting a rich and easy life. They also happen to be the smartphone generation.

So to finish this blog post, let me redefine the generations as follows:

  • The Baby Boomers (too young to remember WWII, and old enough to have built up their wealth before the financial crash): Roughly 1940–1969.
  • The Car-Crash Generation (grew up expecting an easy life, but suddenly the rug got pulled away from under they feet): Roughly 1970–1999.
  • The Smartphone Generation (they don’t remember the easy years, and they live their lives through their smartphones): Roughly 2000–.

bookmark_borderAlphaDiplomacy Zero?

diplomacy game photo
Photo by condredge
When I was still at university, I did several courses in AI, and in one of them we spent a lot of time looking at why Go was so hard to implement. I was therefore very impressed when DeepMind created AlphaGo two years ago and started beating professional players, because it was sooner than I had expected. And I am now overwhelmed by the version called AlphaGo Zero, which is so much better:

Previous versions of AlphaGo initially trained on thousands of human amateur and professional games to learn how to play Go. AlphaGo Zero skips this step and learns to play simply by playing games against itself, starting from completely random play. In doing so, it quickly surpassed human level of play and defeated the previously published champion-defeating version of AlphaGo by 100 games to 0.

It is able to do this by using a novel form of reinforcement learning, in which AlphaGo Zero becomes its own teacher. The system starts off with a neural network that knows nothing about the game of Go. It then plays games against itself, by combining this neural network with a powerful search algorithm. As it plays, the neural network is tuned and updated to predict moves, as well as the eventual winner of the games.

I’m wondering whether the same methodology could be used to create a version of Diplomacy.

The game of Diplomacy was invented by Allan B. Calhamer in 1954. The seven players represent the great powers of pre-WWI Europe, but differently from many other board games, there are no dice – nothing is random. In effect it’s more like chess for seven players, except for the addition of diplomacy, i.e., negotiation. For instance, if I’m France and attack England on my own, it’s likely our units will simply bounce; to succeed, I need to convince Germany or Russia to join me, or I need to convince England I’m their friend and that it’ll be perfectly safe to move all their units to Russia or Germany without leaving any of them behind.

Implementing a computer version of Diplomacy without the negotiation aspect isn’t much use (or fun), and implementing human negotiation capabilities is a bit beyond the ability of current computational linguistics techniques.

However, why not simply let AlphaDiplomacy Zero develop its own language? It will probably look rather odd to a human observer, perhaps a bit like Facebook’s recent AI experiment:

Well, weirder than this, of course, because Facebook’s Alice and Bob started out with standard English. AlphaDiplomacy Zero might decide that “Jiorgiougj” means “Let’s gang up on Germany”, and that “Oihuergiub” means “I’ll let you have Belgium if I can have Norway.”

It would be fascinating to study this language afterwords. How many words would it have? How complex would the grammar be? Would it be fundamentally different from human languages? How would it evolve over time?

It would also be fascinating for students of politics and diplomacy to study AlphaDiplomacy’s negotiation strategies (once the linguists had translated it). Would it come up with completely new approaches?

I really hope DeepMind will try this out one day soon. It would be truly fascinating, not just as a board game, but as a study in linguistic universals and politics.

It would tick so many of my boxes in one go (linguistics, AI, Diplomacy and politics). I can’t wait!

bookmark_borderThe future belongs to small and weird languages

tlingit photo
Photo by David~O
Google Translate and other current machine translation programs are based on bilingual corpora, i.e., collections of translated texts. They translate a text by breaking it into bits, finding similarities in the corpus, selecting the corresponding bits in the other language and then stringing the translation snippets together again. It works surprisingly well, but it means that current machine translation can never get better than existing translations (errors in the corpus will get replicated), and also that it’s practically impossible to add a language that very few translations exist for (this is for instance a challenge for adding Scots, because very few people translate to or from this language).

My prediction is that the next big break-through in computational linguistics will involve deducing meaning from monolingual corpora, i.e., figuring out the meaning of a word by analysing how it’s used. If somebody then manages to construct a computational representation of meaning (perhaps aided by brain research), it should then theoretically be possible to translate from one language into another without ever having seen a translation before, by turning language into meaning and back into another language. I’ve no idea when this is going to happen, but I presume Google and other big software companies are throwing big money at this problem, so it might not be too far away. My gut feeling would be 10–20 years from now.

Interestingly, once this form of machine translation has been invented, translating between two language varieties will be just as easy as translating between two separate languages. So you could translate a text in British English into American English, or formal language into informal, or Geordie into Scouse. You could even ask for Wuthering Heights as J.K. Rowling would have written it.

Also, the computer could be analysing your use of language and start mimicking it – using the same words and phrases with the same pronunciation. In effect, it could start sounding like you (or like your mum, Alex Salmond or Marilyn Monroe if you so desired).

This will have huge repercussions for dialects and small languages.

At the moment, we’re surrounded by big languages – they dominate written materials as well as TV and movies, and most computer interfaces work best in them. It’s also hard to speak a non-standard variety of a big language, because speech recognition and machine translation programs tend to fall over when the way you speak doesn’t conform. Scottish people are very aware of this, as shown by the famous elevator sketch:

https://www.youtube.com/watch?v=sAz_UvnUeuU

However, if my predictions come out true, all of that will change. As soon as a corpus exists (and that can include spoken language, not just written texts), the computer should be able to figure our how to speak and understand this variety. Because translation is always easier and more accurate between similar language varieties than between very different ones, people might prefer to get everything translated or dubbed into their own variety. So you will never need to hear RP or American English again if you don’t want to – you can get everything in your own variety of Scottish English instead. Or in broad Scots. Or in Gaelic.

Every village used to have its own speech variety (its patois to use the French term). The reformation initiated a process of language standardisation, and this got a huge boost when all children started going to school to learn to read and write (not necessarily well, but always in the standard language). When radio was invented, the spoken language started converging, too, and television made this even more ubiquitous. We’re now in a situation where lots of traditional languages and dialects are threatened with extinction.

If computers start being good at picking up the local lingo, all of that will potentially change again. There will be no great incentive to learn a standard variety of a language if your computer can always bridge the gap if other people don’t understand it. The languages of the world might start diverging again. That will be interesting.

bookmark_borderFastern’s E’en and Fastelavn

Today (Sunday 26/02/17) it’s Fastelavn in Denmark. The word comes from Low German vastel-avent, meaning the evening (and by extension the day) before the fast (Lent), which means that it always takes place on the Sunday before Shrove Tuesday.

When I was a kid, we dressed up and went guising, and we took turns beating up a barrel containing sweets with a bat. I’m not sure many kids go guising any more, but the barrel smashing (“at slå katten af tønden”) is still very popular.

Interestingly, in Scotland Shrove Tuesday used to be called Fastern’s E’en, which is clearly etymologically the same word as Fastelavn. It was marked in various ways, but eating pancakes doesn’t seem to have been one of them.

In the Borders it was traditional to have a Baw game:

https://www.youtube.com/watch?v=CID3Y8WN178

Here are a list of the various traditions I’ve managed to find on Tobar an Dualchais:

  • On Shrove Tuesday they had a big bannock with a ring, a sixpence and/or a button in it.
  • Up till about the First World War, a ba [handball] game used to be played on Fastern’s E’en (Shrove Tuesday). People ate currant dumplings on that day.
  • On Shrove Tuesday children went round the houses for bannocks containing a ring or a button. The one who got a ring would be first married.
  • Fastern’s Een [Shrove Tuesday] was celebrated with the baking of special cakes.
  • In the Melrose marriage ball game, which replaced an earlier Fastern’s Een handball game, the bride kicks off a rugby ball in the square and the young men scramble for it. There has been an attempt to stop the tradition because of the danger from traffic. In the earlier game, there were no teams, just small groups trying to run away with the ball. It had to be hidden (not in a house) for three days in order to win the game.
  • On Shrove Tuesday Mrs Hailstones’ mother, who was English, used to make big pancakes and sugar them and roll them up. Scottish families did not do this, but Mrs Hailstones used to do it for her own children.
  • Shrove Tuesday was the day before Lent started. There used to be a big feast on the evening of shrove Tuesday. It was believed that something bad would happen during Easter if Shrove Tuesday was not properly celebrated.
  • The contributor’s great-grandmother (who lived to be 120) was once without a chicken for Shrove Tuesday. A pigeon came in, and so it was killed instead.
  • On Shrove Tuesday children went round the houses for bannocks containing a ring or a button. The one who got a ring would be first married. A button meant an old maid or a bachelor. They had a half day off school on Shrove Tuesday.
  • On Shrove Tuesday a sheep was killed and there was a feast. A broth was made with barley that had been threshed with a ‘cnotag’, a stone with a hollow in the middle.
  • The contributor explains how a bannock was made with barley meal, butter, eggs and sugar, to celebrate Shrove Tuesday. She has never seen it made, but her father saw it in Barra.

It would be nice to see a revival of some of these traditions!