Sorry, I mispoke. It should have been something like “tendency,” as in “users of a language have a tendency to simplify language, but they also have the need to make things clear so they can communicate.” I was just typing off the cuff.
To sort of summarize - there is a tendency among many languages to become morphologically simpler over time. Morphology refers to the habit of building words out of smaller parts; English, for example, has the plural morpheme /s/ (realized as /s/, /z/, or /@z/, for instance, “cats, dogs, and foxes”.) Some languages have lots of morphology, and some have less. Mandarin Chinese, for instance, has very little morphology - no inflections or affixes, really, so the only morphological process that’s important in the language is compounding.
The reason languages tend to become morphologically simpler may be mostly due to phonological change. When a language’s sound inventory changes, those changes are almost always strictly on the basis of sound and not meaning. When a sound change occurs that makes two sounds identical - for instance, the “tap” realization of English /t/ and /d/ when between vowels (“latter” and “ladder” sound identical in casual speech in most English dialects), information is lost; if the merger of /t/ and /d/ in that particular phonetic environment proceeds far enough, English speakers will forget that there ever was a difference between them.
So as sound changes accumulate, they tend to result in loss of information simply because once a merger happens, it can’t very well un-happen. Words will become identical; sometimes that results in homonyms, other times it motivates speakers to find a new word for one concept. When it affects inflections - prefixes or suffixes that indicate a word’s role in the sentence or show agreement with other words - that tends to weaken the inflectional system.
An example: Classical Latin had five vowels, each in both long and short forms, so they distinguished ten different vowel sounds. One of the changes common to most of the Romance languages was to lose vowel length - long and short stopped being distinguishing features for vowels. The various vowel sounds were organized differently in different Romance languages, but the result was that noun and verb inflections - which in many instances depended on vowels - tended to become less distinct. Because of the nature of Latin verb inflection, it was less affected by the change, though all the modern Romance languages have fewer morphological tenses than Latin. But with nouns, so many of the endings became identical or at least very similar that noun cases (cases indicate a noun’s role in a sentence: subject, object, object of a preposition, etc.) began disappearing one by one. As they disappeared, their function was taken over by prepositions in most cases, and by word order: subjects had to appear before verbs, and objects after verbs; an instrument used to accomplish a task was accompanied by “with” rather than being in the ablative case, and so on.
So, since sound changes simply seem to happen inexorably, it’s easy to see why morphological complexity tends to be lost. The point was made that it’s not necessarily simpler to use prepositions or word order rather than inflection, and that’s very much true. More complexity had to arise - rules regarding placement of words, rules for which verbs required their compliments to be preceded by which prepositions, that sort of thing - arose to take the place of inflections.
It’s also been pointed out, and this is worth reiterating, that languages spoken by very small, isolated groups often maintain far more complexity in their inflectional systems than most other languages. Of course, none of these extremely complex languages are familiar to most people - Australian aboriginal languages, Native American languages, and other languages spoken by tiny groups of people tend to have far more inflection than other languages. The upheaval of England as various groups mixed was cited above, and that’s a great example. The same goes for Mandarin Chinese, which has long been the language of aristocracy in China. It has fewer tones than most Chinese languages, and fewer different syllables as well - this may be because it was widely used as a second language by many people.
New forms of inflection arise from words getting mooshed together. French, for instance, has lost almost all forms of verb agreement - unlike the other Romance languages, the verb’s ending doesn’t usually indicate who the subject of the sentence is. Therefore, in French (like in English) the subject has to be explicitly stated, whereas in Spanish or Italian, it can be left out because the subject is clear from the verb form. But there’s reasons to believe that the subject pronouns in French are beginning to be attached to verbs - for instance, people will say, "Mon pere, il dit . . . " ("My father, he says . . . "), and this is becoming a more and more common usage; this and other reasons suggest that these subject pronouns are beginning to be not true pronouns, but prefixes attached to the verb that indicate agreement.
So morphological complexity has been conceptualized as a cycle by some theorists: languages lose their morphology over time as a result of the erosion caused by sound change, but then new morphology continuously arises when words get squished together. That may in fact be the source of all forms of inflection in language.
The reason that it appears that most languages lose complexity over time may be a historical accident - we’re in an era in which large societies are the norm, and thus the tendency of languages with large populations of non-native speakers to lose inflection appears to be the way of things, since all familiar languages have non-native speakers. But at the same time, it’s simply hard to see new forms of inflection arising to replace the old ones; people will perceive words to be separate even though they’re gradually being integrated together and losing their separate identities.
Excalibre, I think between your stance on nuclear/nucular and your post in this thread, you’ve become my new favorite poster. Willingness to explain the evolution of language and to do it clearly – wonderful.
For new forms of inflection, it’s been all downhill since Gutenberg. Modern forms of communication have had a pronounced braking affect on changes in widely-used spoken languages.
Great post excalibre!
<whistling, cheering, holding up lighter, yelling “Encore!”>
(I seem to recall once seeing an example of early stage agglutination in current English, but can’t remember the specfics. Anybody have such an example?)
Thinking of some English words in which differing parts of speech are smushed together into a coherent singular concept:
tit-for-tat
pick-me-up
good-for-nothing
ne’er-do-well
The German-style compound noun can still be considered formed by agglutination, even though presence of such nouns in and of itself isn’t enough to get a language classified as agglutinative. English has scads of these nouns,as well: outhouse, keyhole, buttonhook.
and
It was never clear to me how one separates words. It written languages like English, we put spaces between them. But in actual speech, there’s not really much, if any, separation. Is it entirely subjective, or have linguists developed rules of thumb?
Are you asking how one separates words from one another while listening to speech, or how we decide that “teaching” is one word and not “teach ing”? Both are complicated questions. There’s no single criterion for deciding what constitutes two separate words.
There are some ad hoc rules of thumb … but you make an excellent, well-attested point. This was always something to wrangle with when anthropological linguists were studying languages of people without writing systems.
But native speakers can “naturally” mark off words in their head. It’s not really about pauses during normal conversational speed … there’s some real-time semantic analysis going on in the brain, as well, when you get someone to speak slowly and deliberately.
Not any, even in your native tongue. Your brain is just attuned to picking out words.
Excalibre, you certainly do seem to know about this. Can you vet the Wikipedia artile on Vulgar Latin and let us know if it’s basically accurate as far as anyone can know? It seems to make many of the same points that you did, and also has some surprising tidbids regarding vocabulary. For example, fuego and feu (= fire in Spanish and French) come from Latin focus which meant hearth, or the “focus” of the home.
It should also be remembered that such expressions as snow tires, storm shutters, cellar door, and German Club leadership can just as validly be regarded as compound nouns. The fact the we put the spaces in is just a matter of orthography, but not of grammar. None of the Germanic languages spoken today go as far as STandard German in forming compound nouns, but it’s incorrect to suggest that it’s nearly dead in English.
The “focus” thing is definitely true; as a matter of fact, a lot of very basic Latin vocabulary was lost in almost all of the Romance speaking world. One of the clues that Sardinian is very different from other Romance languages is that it retains words like domus (“house”, replaced by casa, “shack” in other Romance languages). There’s quite a few such basic words that for whatever reason disappeared, or were retained only in Sardinian.
One of the criteria for deciding whether something is a compound noun in English is stress - for instance, “darkroom” is stressed on the first syllable only, while “dark room” (describing a room that is dark, not a place for developing photos) has stress on both words. I would agree with the first two examples you give, but I think I have stress on both words in “cellar door” and definitely in “German Club leadership”, which suggests that for me the latter two are not single compound words.
The point that our own writing system is somewhat arbitrary that way, though, is well-taken; compounds are sometimes written as two words, sometimes with a hyphen, and sometimes as a single word, with relatively little rhyme or reason.
Actually (not to mess with Excalibre too much as (s)he has done such an excellent job explaining a myriad of things already) the room where photographs are developed has primary stress on the initial syllable (dark) and a secondary stress on the ultimate syllable (room) as opposed to a room that is dark which has a primary stress on both words. Another good example of this is:
White House: (primary + secondary)
white house: (primary) + (primary)
It has to do with how sounds in a particular language are arranged - phonemes -> syllables -> prosodic words. *White House * and *white house * are exactly the same with respect to their phoneme and syllable arrangement but they differ w.r.t. their prosodic arrangement (i.e. *White House * is one prosodic word while *white house * is two).
This actually goes back to something that bordelond said. Figuring out a language’s stress pattern is very useful when trying to transcribe an unwritten language. As noted, with morphologically complex languages it is difficult to figure out where the words are. But if you can figure out where the primary stresses occur (and each language has rules which govern this) then you can start to divide up those chunks of sound into “words.”
I’ve read it. It looks decent, though I think there’s some minor details on the sound shifts between Vulgar Latin and Romance that should be either clarified or corrected.
The first book is actually titled The Power of Babel
:smack:
Thanks for the correction.
I see this claim a lot, but always uncited. Is there any evidence for it? It seems like you could just as easily make the opposite claim: that mass communication increases the rate of change of language by spreading new coinages, usages, and pronunciations further and faster.
bit of a hijack, but isn’t Welsh an example of a language that uses prefixes rather than suffixes to express the inflections?