Thursday 18 August 2011

Post Mortem: My 4th Language Creation Conference Presentation

As I've written before, besides organising the LCC4 itself I was also a speaker there (thanks again David for giving me this opportunity to flex my presenting muscles!). While it's probably fair to say that it went well, nothing is ever perfect and I felt it deserved some kind of review.

My presentation's full title was: When Morphology Kicks Syntax Out of the Way, When English Just Seems to Lack the Words, When a Boy Plays with the One That Is Wearing the Beret, and When Over-Long and Obscure Titles Seem Like the Way to Go: an Introduction to Surdéclinaison. As its title indicates, the whole thing was supposed to be tongue-in-cheek in style (although the contents were serious), and I didn't feel I could achieve the necessary lightness using traditional slides. So I went out on a limb and tried for the first time to create a prezi. One of the advantages of prezis is the ability to embed them in webpages, so here it is for your perusal (you'll need Flash for this to work):

Given that it was my first time ever using Prezi, I think I did a pretty good job. Using the ability to zoom in for details and to zoom out for generalities came to me quite naturally. The transitions felt right, as far as I can tell. To prevent my audience from getting motion sickness I added only very few rotations, but that made them all the more effective. Probably my only big mistake was to use a purple colour for the subtitles. I basically had this "brilliant" idea to use the colours of the Conlang Flag in the presentation itself. It felt very right, especially since the prezi itself starts with it. The problem came when I hid small text inside bigger text (a usual effect in Prezi, to create surprise and hide details that only become visible by zooming). I had to use subtitle text for that (the only coloured text style that was available to me), but while it looked fine on my computer screen, I should have realised that purple text on a dark grey background would lack contrast when presented via a beamer. And that it would be colour-blind-unfriendly in any case! Oh well, it's a mistake I'll make sure not to make again.

Apart from this issue, I think I did well to abandon traditional slides and to adopt Prezi as my presentation tool. But I'll let you be the judge of that. Don't hesitate to give me feedback in the comments!
Of course, my presentation was recorded. So if you want to see me make a fool of myself, just watch the video below! (the presentation itself starts at the 05:44 mark)

Seriously, besides my accent (I didn't know it was so bad!), and my propensity to say "eh..." all the time, I think it went quite well. Quite surprising given how nervous I was before the presentation. I guess all that involvement in theatre did pay off! I'm especially glad that people actually laughed, and mostly at the places I was expecting them to! I guess I don't suck as much at humour as I originally thought! Still, some things could have gone better:

  • As soon as I started the presentation it felt like I had lost half my vocabulary. I mean it! I just couldn't find my words (hence the many "eh..."). I blame this issue on stress, lack of preparation, and lack of sleep. I never managed to find time for a dry run. Next time, I really need to get my presentation ready at least a week before presenting, rather than two days like this time. And of course, I need to get enough sleep (says he, while typing this at 2 o'clock in the morning!).
  • I'm not sure I got the message through that I intended. While the presentation was ostensibly about the morphological phenomenon of surdéclinaison, the main thing I wanted people to get out of it was the wonderful variety of linguistic strategies used by speakers all over the world. I wanted to tell naturalistic conlangers: "Beware of the limits you don't even know you're building into your own conlanging. Break the mould, and don't hesitate to question even the most basic assumptions you have about human languages." Oh well, I'm hoping the presentation itself will have rekindled people's conlang bug, and that my Moten won't be the only existing conlang with surdéclinaison for much longer.
  • I made a stupid mistake, when I translated the Basque word galdegaia into English as "topic", when it really means "focus". How could I make that mistake at all?!

Anything else you want to say about my performance? Don't hesitate to give me feedback in the comments.

That's about all I have to say about the form of my presentation. So what about the contents? I personally think I managed to strike the right balance between theory and examples, between jokes and serious matters. However, I'm probably not the right person to critique that part of my work, so I could do with everyone's feedback here.

My presentation resulted in a few questions, and I noticed a few comments on the IRC chat log worth answering to. So let's use the remainder of this blog post to answer people's questions, remarks and concerns. If you didn't attend the LCC4, you should really go through my prezi and watch the presentation or what I'm going to write below won't make much sense to you.

After the presentation, Tam Blaxter came to me to ask whether group marking case suffixes wouldn't be better described as clitics rather than true suffixes, in which case their behaviour is described as syntax rather than morphology, and recursion phenomena like surdéclinaison become less exceptional. This is actually a very good question, in that it describes well the currently accepted truth among the English speaking linguistic community, and the main reason there's hardly any talk of surdéclinaison in English linguistic literature: English speaking linguists generally don't believe it's an actual morphological phenomenon, but just a case of clitic stacking, which is hardly exceptional. To this question, I actually have two answers:

  1. I often feel that this is a cop-out: every time someone uncovers a morphological phenomenon that doesn't fit people's expectations of what morphology should be like, they just shout "clitics!" and somehow feel they have solved the issue. I call that hand waving (not that I believe Tam was guilty of that. I know he meant the question as an actual request for information).
  2. However, there is such a thing as clitics, and indeed declension paradigms often do originate from grammaticalised clitics. And since hand waving away a hand wave is still hand waving (following me so far?), it's actually worth answering that question, at least for Basque. To do so, just take a look at the table below. It's a full paradigm of the word etxe: house. The exact meaning of each form is irrelevant here, just focus on how they look like. The Basque noun paradigm is based on two elements: the definite article (-a in the singular, -e in the plural, the indefinite, without article, being undefined for number) and the case suffix, which follows the article when it's present. However, when you look at the actual paradigm, you see that the combination of the article and the case suffix is not always that straightforward. We've got here examples of:
    • Suppletion: the absolutive plural is formed with the unanalysable suffix -ak, rather than -e.
    • Conditional epenthesis and elision: the dative -i and the cases starting with -e- both take an epenthetic -r- when added after a vowel, including the singular article -a, but not after the plural article -e. Moreover, the plural article actually disappears before the case suffixes in -e-, so that the only distinction between the indefinite and the plural is the presence or absence of the epenthetic -r-.
    • Other weird behaviours: the local cases are a weird bunch. Unlike the other cases, adding them to the stem directly doesn't form the indefinite but the singular (except the inessive -n which does take the singular article). To form the indefinite, one has to add -ta- between the stem and the ending. And even more strangely, this -ta- element must also appear between the plural article and the local case ending!
IndefiniteSingularPlural
ABSetxeetxeaetxeak
ERGetxeketxeaketxeek
DATetxerietxearietxeei
GEN.POSetxerenetxearenetxeen
COMetxerekinetxearekinetxeekin
BENetxerentzatetxearentzatetxeentzat
CAUSetxerengatiketxearengatiketxeengatik
INSetxezetxeazetxeez
GEN.LOCetxetakoetxekoetxeetako
INEetxetanetxeanetxeetan
ALLetxetaraetxeraetxeetara
TERMetxetarainoetxerainoetxeetaraino
DIRetxetarantzetxerantzetxeetarantz
DESTetxetarakoetxerakoetxeetarako
ABLetxetatiketxetiketxeetatik
PARTetxerik
PROLetxetzat

So, does that mean that the case endings and articles are actual suffixes rather than clitics? I contend that it does. I wouldn't expect a clitic-based paradigm to be that complicated. Also, I would expect a clitic-based paradigm to have purely phonological alternations, whereas there is something here occurring at a morphological level. And in any case, I would never expect suppletion to appear in a clitic-based paradigm! All in all, this makes me conclude that the Basque case endings are true suffixes. You may disagree. In that case, I'll just point out that Moten is definitely group marking and has surdéclinaison, while its core cases are marked by infixes!

On IRC, Philip Newton gave me probably the best compliment one could have given me about my presentation. I'll just quote what he wrote:

11:39:18 AM pne: I think Christophe did an excellent job of knowing his slides - he didn't have to look at the screen to know where he was or what he was going to say next (more than once or twice)
11:39:41 AM pne: as if he had practised the talk a few times so he knew the order well

I'm honoured by this compliment, especially since it's completely off the mark: as I wrote above, I finished the presentation during the night from Thursday to Friday, and never got the chance to rehearse it. Personally, I believe it looked like I knew my slides very well for two reasons:

  • I defined the transitions between the different areas of my Prezi very clearly, and thought in advance about how I would handle those transitions and what I would actually say during them, while I was building the presentation. Thanks to this way of organising myself, I already knew what I would say when going from slide to slide without having to rehearse. It's much easier to improvise between transitions than to improvise the transitions themselves, and getting the transitions right does make the whole presentation feel smoother and slicker, which is probably what prompted Philip to make his wonderful comment.
  • As Fenhl wrote after Philip's comment, I did feel that the way Prezi handles the presentation of information was very natural to me. It somehow "clicked" with me much better than the traditional Powerpoint style of presentation.

Also on IRC, Broca had a general complaint during my presentation that examples were not all accompanied with interlinears. In theory, I agree with Broca: especially when using conlangs, which can be quite exotic, to illustrate some linguistic phenomenon, which can be quite obscure, the only way you can make sure people understand how the examples work is by showing how they are constructed, using interlinears. However, as usual there is quite a difference between theory and practice, and there are legitimate reasons to not use interlinears with every example, at least in a presentation:

  • Interlinears are heavy. Because of the use of dashes, dots and abbreviations added on content words, interlinears can easily become twice as long as the example they represent, if not more. That can be a lot of lost screen estate in a presentation, possibly forcing one to use a smaller (and thus more difficult to read) font. Also, with too many interlinears on a slide, you can get the effect of missing the forest for the trees: they go against the principle that one should use as little text as possible in a slide.
  • When one is illustrating a single linguistic phenomenon using a series of examples in a presentation, interlinears are only useful on the first of them, or the first two at most. There is nothing to gain (in terms of getting a message across) by burdening all the examples with interlinears, unless you want to show phenomena like suppletion or alternative constructions.
  • Interlinears work best when the original contents are aligned with the interlinear. The drawback is that the original example becomes more difficult to read because of the added spaces. Repeating the example with natural spacing before the example above the interlinear only makes things even heavier.

For this reason, I made the very conscious decision to not include interlinears for all examples. I just added them where I felt they would help getting the message across, and I didn't add too many of them in order not to burden the presentation. Now whether I added enough of them to correctly illustrate my points is not something I can judge by myself, so any feedback about this issue would be more than welcome, so that I can do things even better next time!

Finally, during my talk, I mentioned Jacques Brel's song Rosa, a song whose chorus is the full declension of the Latin word "rosa", as taught to French-speaking schoolchildren even to this day. While I was doing this, Thavernes on IRC thought they should correct me by pointing out that the song was from Salvatore Adamo, not from Brel. Well, they were wrong. The song Rosa was written by Brel himself (both music and lyrics, both in French and in Dutch) and first released as part of his album Les Bourgeois in 1962. Adamo has indeed covered this song, as he has done for various songs from Brel, but he has not written it. Indeed, in 1962 Adamo had yet to release his first album! I know it's a tangent but I just wanted to make sure the correction would get out.

Well, that's about all I have to say. This post has been very long in the making (I started working on it in back in May!), so my apologies to those who have been waiting for it. I hope you'll enjoy it, and that it will be useful to those who missed my presentation as well as to those who saw it. As I wrote various times, I'm dying to get some feedback, so don't hesitate to use the comment area below. And as a special request, if you know of any conlang that uses surdéclinaison (actual surdéclinaison, not suffixaufnahme!), or have created one, please give me a big shout! I'm dying to see how others would implement this feature.

No comments:

Post a Comment