This doesn't sound at first like a psycholinguistics topic, but the answer is surprisingly related to the way our brains process spoken language. This post is also 80% inspired by a 2010 article in NIPS, which happens to be one of my favorite scientific papers of all time. It'll be given in the references of part 2, and although it's pretty typically dense academic writing I strongly recommend reading it. It's a pretty unparalleled paper and it builds on some fundamental concepts related to this question. But first, some groundwork:
So you might notice that when forming straightforward transitive sentences in English we generally employ Subject-Verb-Object order, as in "John eats bagels." Unsurprisingly, not all languages employ this basic order - there are six possible combinations (3!, all possible permutations of S, V, and O), all of which have been claimed to dominate in at least one language. Perhaps surprisingly, some basic word orders are vastly more common than others. In general, SOV and SVO are most common, together accounting for at least 90% of languages with a default word order, and roughly as common as one another. VSO, depending on who you ask, claims around 10%, it can be said for VOS that it at the very least exists, and there are no uncontroversial instances of object-initial orders OSV and Klingon's OVS. This distribution leaves some explaining to do.
Leaving aside some fanciful early explanations, the most substantially agreed-on explanation of this involves three principles first envisioned by Joseph Greenberg and laid out in 1986 by Russell Tomlin. Two of them, the "Theme First" and "Animated First" principles, have essentially the same effect. Across languages, people talk in essentially the same ways about the same things. We care about people or other agents capable of influencing the world around them, and tend to focus on individuals as our main cast throughout a discourse or story. Animate, anthropomorphic agents whom we cast in the starring role of our strings of sentences are overwhelmingly grammatical subjects. If they are unfortunate enough to be the patient rather than the agent of a verb, befallen by the world acting on them rather than the other way around, we're most likely to passivize the clause, like "Rob was hit by a car," rather than give a newly introduced figure the honor of playing subject. This is because a subject is not primarily the designation for the doer of a verb as we learn in fourth grade but rather of the topic of discourse. When Mrs. Miggins told you in English class that the passive was unnecessarily wordy and weak, she was wrong - it's actually a very handy tool for keeping the people we care about at the center of the action. In topic-prominent languages like Japanese, where a topic is a grammatical role divorced from agentivity, the topic almost always comes first. Similarly, in languages like Navajo where word order is based on animacy, more animate actors like people and lightning come first. Because animates and themes are dragged to the front, and are so ubiquitously muddled together with the designation of subject, subjects are very likely to be the first nouns in the sentence. Indeed, languages where subjects come after object by default barely make it onto the scoreboard. The last principle, "Verb-Object Bonding," simply dictates that verbs and objects stick together, probably because they so often form a complex called the predicate. VSO's only crime is splitting the two beloveds apart, and while that doesn't remove it from existence it certainly seems to make it less popular.
This still leaves the "Theme First" and "Animated First" principles, which dominate our syntactic rules, unexplained. I know I promised psycholinguistics, in part two I'll introduce a potential why that has big things to say about brains. This topic is one of my favorites in linguistics, and I had so much fun writing about it I didn't even make it to the best part within 500 words! Catch up in two weeks to find out what this has been building up to.
Further reading:
Word Order
WALS Word Order Map
Tomlin 1986 behind a paywall :(
Flamboyantly colored overview of Tomlin's findings
Languages with unusual word orders:
Mayan
Malagasy
Hixkaryana
Xavante
No comments:
Post a Comment