April 2008
Sun Mon Tue Wed Thu Fri Sat
    1 2 3 4 5
6 7 8 9 10 11 12
13 14 15 16 17 18 19
20 21 22 23 24 25 26
27 28 29 30      
Mar   May


Blog-Parents

RaptorMagic

Orcinus

Blog-Brothers

Callimachus
(Done with Mirrors)

Gelmo
(Statistical blah blah blah)

Other Blogs I Read
Regularly Often

Athletics Nation

Andrew Sullivan
(Daily Dish)

Kevin Drum
(Political Animal)

Hilzoy
(Obsidian Wings)

 Friday, April 18, 2008
Whose Bounty?

This began as a response to a discussion on our brother blog Done with Mirrors. That discussion is a week old now, and my response is a bit long for the comments box anyway, so I'm posting it here instead. In brief, Callimachus read an opinion column in the L.A. Times offering sundry etymological and political observations centering around the way the Arabic word "Allah" is used in English. Cal perceived this as an effort to impose change on the English language for political ends, and he objected accordingly.

The article is a pleasant read, but my response is to the DwM discussion. Among the comments, most concur with Callimachus, with several further arguing that Allah really isn't the same person as God and it's outrageous that someone should try to manipulate language in order to trick us into thinking he is. One (Michael J Totten) is sympathetic with Cal's objection to what they perceive as the "p.c." tone of the Times column, but emphasizes that the author is entirely correct on the central factual point: that "Allah" in Arabic really does just mean "God", and it isn't a proper name for their specific god as it in English. (Michael has spent a great deal of time among Arabic speakers, so he ought to know.)

Like Callimachus, I have little interest in the theological question of whether the Christian and Muslim gods really are the same being. He and I are both more interested in the linguistic question, and on that point I think he's got it wrong. Cal argues that the English language is democratic and oughtn't be subjected to decrees imposed from above. So far, I agree (and I would add that such decrees never work in English, anyway, no matter how hard pedants try). But I don't think the Times columnist is suggesting that. In a democratic system -- which we agree the English language is -- change comes from changing public opinion, and the column is intended as a step toward that end. And I think it's right on the point.

Failure to Translate

If you're really going to pursue this question, I think you have to look at the history of the words in question. That "Allah" suggests a separate deity in English in a way that "Dios" does not is the result of centuries' tradition of translating other words that mean God while neglecting to translate "Allah". This is a choice we have made as a culture (or rather, a choice our predecessors have made) to deny the equivalence of their god with ours.

No, the meaning of the two words isn't exactly identical, but it never is in translation. Rojo doesn't mean exactly the same thing as red, either, but that doesn't stop us from making the obvious translation. Russians don't mean exactly the same thing by "Bog" that we mean by "God", yet we don't say that Russian Orthodox Christians worship Bog. We say they worship God. When we (historically) say that Greeks worship God, Russians worship God, Armenians worship God, Jews worship God, but Arabs worship Allah, we are deliberately defining an extra level of exclusion from certain faiths.

And of course it's easy to see the history behind that. Most of my readers aren't medievalists, but I think the essential point is familiar enough. Central to the formation of European cultural identity was the idea of opposition to Islam. Even in the 18th and 19th centuries, when many came to romanticize the Islamic East, it was still done in a way that defined the East as "other".

History aside, two basic linguistic points stand out: (1) writing "Allah" as "Allah", rather than translating it to "God", is anomalous in comparison with how we treat corresponding words in other languages, and (2) doing so is (was) a political decision.

There's no non-political basis for it. In the evolution of the belief systems, there is just as much connection between the Christian and Muslim ideas of God as there is between the Christian and Jewish. Etymologically the words are connected, as we see in "hallelujah". Ask for the literal meaning of "hallelujah" and you'll be told that it's "praise be to God", not "praise be to El". Ask for the literal meaning of "allahu akbar" and you'll be told that it is "Allah is great", not "God is great".

Some of our servicemen in Iraq have picked up a curious Arabic phrase, "inshallah". It is a frequently used idiom -- not just in Arabic, but in Persian, Turkish and Urdu as well -- and depending on context and tone of voice it can express any in a wide variety of emotions, including hope, desire, or fatalism. If you are asked for the literal meaning, you have a choice. You can say it means "if God wills it", or you can say "if Allah wills it". With the former, you have the beginnings of an interesting discussion of the rich connotations of this fascinating phrase. With the latter, you suggest a competition between Allah's will and anyone else's will, which is contrary to how the phrase is used in actual conversation.

Communication

Theodore Bernstein, legendary editor of the New York Times in the middle of the 20th century, wrote several books on style and usage. Like all serious grammarians, he often wrestled with the question of when a change in usage should be resisted and when it should be embraced. In such cases, Bernstein's touchstone was always effectiveness of communication: Does this change result in writing that communicates better or not as well?

The reason I object to treating the Arabic word "Allah" as a proper name, rather than translating it to "God", is not that it's inconsistent with how we treat other languages; it has nothing to do with whether I believe our god is the same entity as their god, nor does it come from any desire to appease anyone politically. I object to it because, as an inaccurate translation, it hinders our understanding of what was actually said.

The reason we translate from other languages is because we want to know what others are saying. We want to share in whatever knowledge the literature of other languages might have to offer. Whenever we mistranslate, it is our loss.

What would Pascal's wager mean to us if his argument was that one must bet on the existence of Dieu? How less meaningful would Dostoyevsky be to us if he were contemplating the idea that if Bog didn't exist all things would be permissible? These are ideas about God that are universal. By failing to call God by a name that means something to us in English, we separate ourselves from the concept, and our understanding is diminished.

The slogan that has inhabited the top of this blog for the past three and a half years is a saying attributed to the prophet Muhammad:

Difference of opinion among my community is a sign of the bounty of God.

No doubt the Prophet spoke in Arabic, and no doubt he used the word "Allah". When I write it in English, I write "God". If I were to say "bounty of Allah" it would absurd. It would be like I'm claiming the approval of a foreign deity. Although the source of the maxim is also interesting to me, I didn't choose it because it was an Islamic saying. I chose it because I agree with the message, and I believe that message is universal. If I were obliged to write "Allah" instead of "God", it would deny this bit of wisdom to non-Muslims.

Postscript

Getting back to the newspaper column that started this discussion, I must say that while the columnist is absolutely right in recognizing that "Allah" is misused in English, his conclusion is dead wrong. From the paragraph Callimachus quotes:

The two words, "God" and "Allah," do not mean the same thing in English. They should.

No, they shouldn't. The word "Allah" shouldn't be in the English language at all, any more than kami or Mungu should. It's not an English word. It's an Arabic word. And it means "God".

11:51:41 PM  [permalink]  comment []