Saturday, October 31, 2009

The "Indo-European Family"

The other day, I wrote:
For at least fifteen years now, I've taken an interest in the Indo-European family and the relationship between older languages. Sometimes, this is just for curiosity's sake. Sometimes, it helps me better understand modern languages.
Typically, when you’re given a list of Indo-European cognates, the name of the game is to show the obvious interrelation between the languages with transparently similar words. Today, I’d like to take a different tack, showing where different words come from and how an understanding of the Indo-European background can offer better understanding and help you make connections you might otherwise miss. The topic is family vocabulary; the languages we’re looking at are Irish, Italian and Dari. To make the connections for Italian and Irish, I’ll be going by way of Latin and Old Irish. I don’t happen to know anything about Avestan, Old Persian, etc, so we’ll have to make some logical leaps on the Dari.

Father
Proto-Indo-European pəter > Latin pater > Italian padre (note that the “ə” denotes a laryngeal, i.e. a sound in the throat that developed into a vowel later)
PIE pəter > Old Irish athair > athair (initial “p” in PIE and proto-Celtic dropped by the time of Classical Old Irish)
PIE pəter > Dari padar

Mother
PIE mater > Lt. mater > It. madre (works just the same as pəter)
PIE mater > Oir. máthair > Ir. máthair (same)
PIE mater > Dari madar

Mother and father are pretty easy. There’s a third word in the same family:

Brother
PIE bhrater > Lt. frater > It. fratello (diminuitive suffix to mark a little “frater”)
PIE bhrater > OIr. derb (certain) + bratháir > derbh-bhratháir > Ir. deartháir (imagine if in English we said “dear brother” so often that we started running it together as “dearother” and you can sort of see what happened.)
PIE bhrater > Dari bradar

The next word has a rather different base form, but notice how Italian and Irish make the same transformations to it as with brother:

Sister
PIE swesor > Lt. soror > It. sorella
PIE swesor > OIr. derb (certain) + siur > derb-ṡiur > Ir. deirfiúr
PIE swesor > Dari khahar (cf. Middle Iranian khwahar, with PIE s > h/kh; this also happens with Greek, hence Grk. helios, Lt. sol as in solar)

Our last two words teach us a different lesson, namely that even when modern Indo-European languages diverge in vocabulary, knowing the background of the words can help us make new connections.

Son
PIE dhei- (to suckle) > Lt. filius > It. figlio (think of filial devotion)
PIE maghu (youngster) > OIr. macc > Ir. mac (maghu is also the source of maiden)
The Dari is pisar, cf. Farsi pesar, Lat. pes, Grk. paidos; not sure of the PIE root

Daughter
PIE dhei- (to suckle) > Lt. filia > It. figlia (feminine of figlio)
Proto-Celtic eni-gena > OIr. ingen > Ir. iníon (presumably, a daughter is “born into” a clan)
PIE dhugeter > Dari dukhtar (that one was pretty easy!)

Obviously, you wouldn’t want to do this kind of research with every word you ever intend to learn! On the other hand, trying to come up with a mnemonic for every word can get pretty tedious too. The important thing is to find devices for better remembering and understanding what you learn so that it will stay with you. And sometimes etymology is just what you need to make a connection (a real connection!) that you might have missed. This is especially the case if you use it often, as you will start to develop an intuition for connections between languages and language families so that as you learn new words, they come alive on the basis of old patterns you’ve already figured out.

By the way, if you suspect a relationship between an English word and a word from another Indo-European language, a great place for etymologies is etymonline.

Saturday, October 24, 2009

Dari and Irish

Last week, I mentioned that my plans for Indo-European were to sort of fill in the pieces of the puzzle as I went along. This week, I’ve been working at nearly opposite edges – a couple lessons of Pimsleur Dari and some dabbling in Irish.

Dari
Dari is one of the two main languages of Afghanistan, the other being Pashto. Dari is pretty closely related to Farsi – indeed, it has been called Kabul Persian. If you already speak some Farsi, Dari is pretty easy – much of the vocabulary is similar, if not identical, and the basic grammar is not that different. The main challenge is keeping track of the different locutions, like remembering whether it’s Farsi that asks about your health while Dari just asks how you are or the other way around (it’s the first, at least according to the Pimsleur programs).

Pimsleur Dari is pretty much like all the Pimsleur programs. If you’re going to Afghanistan or just curious about the language, the Pimsleur Basic program’s a good place to start before moving on to more in-depth resources. And if you’re interested in how regional dialects or sister languages can do different things with essentially the same words and structures, comparing and contrasting with the Basic Farsi program is a fun little exercise.

Irish
For Irish, I’ve been listening to Pimsleur’s Quick and Simple program. The first thing that strikes me is just how much the word maith, good, comes up. From “thank you” - go raibh maith agat - to “I would like…” – ba mhaith liom - it seems there’s no end to things being good for you or me, or maybe not so good. The other thing, for which I’ll congratulate Pimsleur, is the ease with which it introduces lenition and eclipsis by way of sentences about understanding. Here are the main sentences used:

Main form: Tuigim (I understand)/Tuigin tú (you understand)

Lenition: Ní thuigim (I don’t understand)
Lenition is a softening of the consonant. In this case, the initial “t” softens to “th,” pronounced “h.” We soften consonants in English too, for example, pronouncing “water” as “wadder” when we’re talking quickly. The tricky thing about lenition is that you usually have to go back to Primitive Irish or Proto-Celtic to find a word ending, now lost, that changed the phonetic environment such that the first consonant of the lenited word got softened.

Eclipsis: An dtuigin tú? (Do you understand?)
Eclipsis means that you add another consonant to the beginning of the word, and that consonant eclipses the first one. In this case, the “d” eclipses the “t” so you pronounce the second word “diggin” with a “d”. In Old Irish grammars, eclipsis is usually referred to as nasalization. This is because eclipsis/nasalization tends to occur when the preceding word ending in “m” or “n” in Proto-Celtic or Primitive Irish. That “b,” “d” and “g” are eclipsed by “m,” “n” and “n” respectively – literally nasalized! – while the unvoiced stops are eclipsed by their voiced counterparts should give a sense of how eclipsis is not really about eclipsing the second consonant so much as adding a nasal character to it to make up for the preceding nasal that has been lost.

How do Irish speakers keep track of all this, by the way? Automatically, I imagine. Little French kids can learn that it’s “des yeux” (day zyeuh) but “des voitures” (day vwah-tur) well before they can read and write; presumably, Irish speakers, too, are aware of their “endings” that change the next word, otherwise initial mutation would have died out long ago.

Now, here’s the nice thing about Pimsleur: They don’t teach you about lenition, eclipsis, the history or Irish or any of the stuff above. They just teach you, as they come up, which words cause the next word to change and which change is caused. When you contrast “go maith” and “an-mhaith,” for example, there’s little fuss about theory, just a note that “m” softens to “v.” Since it’s a short course, they can get away with this, of course, since you only need to know that there’s a system here, not all the details for how it works.

I imagine if Michel Thomas were teaching it, we’d have the zero track – no mutation, the “h” track – words that cause lenition, and the “2” track – words that cause double consonants…

-Now how do you say, “I understand”?
-Tuigim.
-Good, and how do you say, “I don’t understand.”
-Ní tuigim?
-No, no, no, you have to get on the “h” track. Remember, “t” goes to “th,” pronounced as…
-“h”… Ní thuigim.
-Exactly!
For Irish, I’ve also been using Living Language’s Spoken World Irish. It’s not a bad program. You start with key vocabulary, then get a short dialog with a translation, then an introduction to some grammar points, followed by exercises. The audio on the CDs for use with the book is good. There is also a second set of CDs to use after you’ve finished a lesson for review in the car or whatever. These are basic enough that you can follow along and review key points without needing so little concentration that they’re pointless or so much that you couldn’t really use them on the go. My one complaint: the grammar sections are a bit unfocused. It would be more useful if they were either 1) more comprehensive or 2) more specifically geared to only explaining the grammar so far shown in the dialogs. That said, this is a pretty good course and looks a lot less painful than most of the other stuff I’ve seen. If you want to teach yourself a bit of Irish and want to read and write, not just speak, this is where I’d start.

Saturday, October 17, 2009

Language Goals Update

Earlier today, I was at Omniglot and happened upon his My Languages page. I thought that was kind of a nice idea, so I made one of my own. It's a bit more tedious, but should do the job. For the moment, you'll find it in the post below this one. It's also linked on the sidebar though and I'll be updating it (very intermittently) just in case anyone should ever stop by in search of a bigger picture view of what I work on, as opposed to what I've stumbled upon and thought I'd blog about in the last four or five posts.

Speaking of blogging about things, just a quick plug for wiktionary at en.wiktionary.org: I've been using this to check pronunciations while fussing with Irish and I've been pleasantly surprised at how much it has. If you're studying an unusual language, have a look.

Language Goals

Regular visitors to this site will note that some of my language efforts are a bit scattershot. I've never been a very linear person, so my learning efforts both toward individual languages and toward polyglottism most often resemble filling in a puzzle, not completing a project.

This post, linked in the sidebar, keeps track of what I am currently working on. It is subject to revision as my goals and approaches change.

At the moment, I have two big language goals, partially intertwined:

1) Modern Languages as an access point to the world
Both in my work and in my reading I encounter a number of languages and cultures. Starting with English and French and moving outward in Western Europe, I am trying to make more of the world familiar and accessible to me following on the map below:


Black: Fluent to Highly Proficient: English and French
Dark gray: Fair reading and speaking competency: Spanish, Italian, German
Lighter gray: Basic communication skills: Portuguese
Light gray: Basic survival skills: Greek, Farsi, Dari, Turkish and Uzbek

Note that the map does not show the world: Knowing English, French, Spanish and Portuguese pretty much gives you the Western Hemisphere. It's heartening to see, when contemplating how far these languages will take you, but also a bit gratuitous.

It will be noted that most of my modern languages are Indo-European. The exceptions are Turkish and Uzbek, both Turkic. Other languages, including the dread Mandarin, pop up on my radar screen. But they're not part of my core focus.

2) Indo-European
For at least fifteen years now, I've taken an interest in the Indo-European family and the relationship between older languages. Sometimes, this is just for curiosity's sake. Sometimes, it helps me better understand modern languages. While my interest in "modern languages" is as an access to the world and its cultures, my interest in Indo-European is rooted more in seeing how languages interrelate and evolve. Right now, there are four areas I'm exploring:
  1. Celtic: Old Irish vs. Latin; Old Irish as ancestor to Irish; Irish vs. Breton
  2. Romance: reviewing Latin; reviewing Old French
  3. Germanic: reviewing Old English
  4. Greek: reviewing Classical Greek with a peek at Homeric

Most of these explorations are by way of 1) The Early-Indo-European Online site (linked at right) and 2) Fortson's Intro to Indo-European. My Celtic stuff is supplemented with Stifter's Intro to Old Irish and various resources for modern Irish and Breton.

Plans:
Much to my dismay, nobody is paying me to study this stuff. It's mostly for my own satisfaction. Consequently, my studies are geared to what pleases me. It is my long-term aim to develop a reasonable degree of confidence that if dropped anywhere in the sweep of territory on my Modern Languages map, I'll at least be able to acquire a hot meal, a bed and, if necessary, a consular official who can arrange my safe departure. Among the light grays, I'm actually there with Turkish, Uzbek and Farsi. As for my Indo-European interests, that's a more progressive thing. People say that Italian and Spanish are so close that if you know one you can easily learn the other. The same goes for Spanish and Portuguese. If you know what to look for, you can easily add French. And if you've got a handle on these, a lot of the Latin they came from will look familiar. I'm finding the same thing happens when you go back further. So while I don't intend to learn all the ancient languages any time soon, it's fun to put things together and I'll continue to do so off and on.

Friday, October 16, 2009

Putting Language Pieces Together

I've been looking at modern Irish, Old Irish and proto-Indo-European. Not speaking Indo-European, I tend to use Latin as my cross-reference for thinking about how Old Irish works.

It turns out I'm not the only one baffled by Celtic languages at times. When the Indo-European family was first being considered, they weren't even sure Celtic belonged given its exotic verbal system and the peculiar phenomenon of initial mutation. It was Franz Bopp who realized that Celtic initial mutations were, in a manner of speaking, tacking classic IE endings on the start of the next word. He made the leap in comparing Celtic phrases with Sanskrit equivalents, incidentally.

Now, today, I stumbled upon another nifty bit in Fortson (Intro to IE). In Old Irish another feature you run into is a sort of umlaut - internal vowel mutation - which takes the form of "raising" and "lowering." Again, at first glance it looks like another feature the Celts came up with to confuse everyone else. On closer inspection, though, you find out that the changes take place depending - again - on the missing ending from IE and proto-Celtic.

The only drawback about these neat discoveries is you need to have some familiarity with another older IE language to make use of them. Fortunately, I do, so it's been interesting to see the confusion that is Old Irish slowly start to make sense.

This is one thing to be said for learning multiple languages, or even studying them at the same time. While there's more to learn, there are also more connections to be made.

Friday, October 09, 2009

Odds and Ends

A few notes, and a little housekeeping...

Learning Languages Inside Out
A few weeks ago, I mentioned the Early Indo-European Online site. I've been fussing about a bit with Indo-European of late, using this site, Fortson's Indo-European Language and Culture and Clackson's Indo-European Linguistics. In particular, I've been looking at Old Irish, which has given me occasion for a few a-has and one big duh:

In the Celtic languages - at least the insular ones - there is an odd feature called initial mutation. This means that the beginnings of words change in certain circumstances, not just the ends. One of the features in Old Irish is "nasalisation," which changes the pronunciation of some consonants as well as causing following words that start with a vowel to instead start with the letter "n". This all seems rather bizarre, but then there came the "duh" moment in which I saw that this isn't necessarily just something the Celts came up with to confuse the hell out of the rest of us.

Looking at the declension table for o-stem nouns, I saw that the forms that cause nasalisation are, by and large, those that end with "m" in Latin, "n" in Greek and, surprise, "n" in postulated proto-Indo-European and Proto-Celtic forms. Nasalization, then, is sort of like the French liaison - the ending sound of a word moving to or affecting the beginning of the next ("les yeux" sounds to be pronounced more like "lay z-yeuh" than "layz yeuh", for example). [Commenter Jared Grubb points out that we do this in English too, as in "a... n-octopus."] Continuing to look through the tables, it soon became apparent to me that while I wasn't ready to systematize the relationship between Latin endings and the initial mutations caused by the different Old Irish forms, the Old Irish table no longer looked quite so bizarre.

I call this finding of patterns a "duh" moment because I should have been more alert to there being some method to the madness of Insular Celtic initial mutations, even if I haven't yet sorted out exactly what it is. Language has its own logic; it's knowing where to look for it that is the problem.

[New material follows from the original post...]
Sometimes, to make sense of a language, you have to look at it from a different vantage point. If you can absorb the forms and just be happy knowing, that's great. But if there's something that just baffles, sometimes the history of the language, an analysis of a sentence that's bothering you with a derivation tree (scroll down to Context Free Grammar in this article to see a derivation tree) or comparison with a related language can help you better learn and remember the point because it's not so bizarre once you see what's behind the language working that way. Call it "learning a language inside out".

[Note: This section has been edited, revised and expanded as the initial version stumbled off before I got to what I actually meant by "learning a language inside out."]

Learn a Language While You Sleep?
Only in your dreams. Tim Ferris had a post on lucid dreaming the other day, and one of the one of the "related posts" was language learning. Bearing in mind that lucid dreaming, the sensation of out-of-body-experiences and the feeling that what's happening in your guided meditation is real are all sort of interconnected, I can see using lucid dreaming, if you're good at it, as a way to get comfy with the idea of going to a foreign language environment. But as far as I know, you can only dream about things you either know or have the knowledge to conjecture, so if you're interested in "sleep learning," your best bet is still to listen to vocabulary tapes right before you go to sleep. After all, if you're having a lucid dream, you'll probably be tuning out whatever is on the headphones or, at best, incorporating it in a way that your making mind might not find so useful.

Housekeeping
I've cleaned up the links at the side a little bit, taking out a few sites that haven't updated in a while. If one of those sites is yours and, darn it, you were going to resurrect it later this month, drop me a note in the comments when you start posting again.

Sunday, October 04, 2009

Not so revolutionary Language Revolution

I picked up Tony Buzan's Language Revolution - Spanish more with an eye to whether its methods could be adapted to more exotic languages (Dari and Old Irish - don't ask) than for learning Spanish since it's a beginner book and I already mangle basic Spanish just fine. What I would say, first, is that it looks better in the bookstore when you're thinking about buying it than it does at home when you sit down to work through it. It's not that it won't work, just that there's not much there. If you're planning on a trip and want some nouns and adjectives to toss out there to get what you want in some everyday situation, this is great. It might be really great if you're one of those people who can't pick up those basics by scanning a phrase book. However, it teaches vocabulary, not Spanish.

The biggest disappointment, however, comes in the marketing. This is sold as Language Learning meets Mind Mapping. The concluding paragraph of the introduction includes the following:
You will be able to modify and expand all the Mind Maps in the book to help you test and reinforce your learning. First you will need to download and install the iMindMap(TM) software on your computer.
Maybe I'm reading deficient, but when I stumble over those sentences in a book that sings the praises of its website support, I get the impression that you can download iMindMap(TM) software, or at least a scaled down version of it, at the website. Otherwise, what would would be the point of a language book that teaches you to think creatively about learning languages using the author's mind maps as a springboard to forming your own associations? However, so far I've only discovered a link to another Tony Buzan site where you can download the iMindMap(TM) software for a free trial. I haven't investigated that yet; will do so later. If the Language Revolution site has a special version of the software for users of the book, they've hidden it well. This makes it feel, frankly, like what purports to be a revolution is just another gimmick for driving you to Tony Buzan's other products.

Collins makes some nice stuff and Tony Buzan has some good ideas. But this particular package oversells itself. With a scaled down version of the software easily located at the Collins website, I might have gotten hooked enough to try to the full-scale product. Directing me off-site for a trial of a tool the promotes as one of the great things the book's web support has to offer just makes me wary of the whole package.

I'll be giving the whole thing a little bit of a look and if it seems to have more to offer than what I'm sensing here, I'll post on it. But if you're planning to go to Spain for a week and just want some words and phrases, I'm going to recommend Week One of Speak in a Week and Elisabeth Smith's One Day Spanish as a better way to go.

Info on the book and the reservations I've expressed is welcome in the comments; note that comments are moderated and may not be approved for a few days.