Latin is characterized by a highly elaborate system of noun declensions and verb conjugations. Latin-derived Romance languages have the verb conjugations, but all of them, AFAIK, have dropped the noun declensions, and determine a noun’s part-of-speech by syntax alone. Old English once had a declension system – at least, for pronouns – but it faded with time. http://en.wikipedia.org/wiki/Declension_in_English Why does the evolution of languages appear to tend towards simplification of grammar?
I’m not sure it’s simpler. By dropping the declensions, you make necessary strict rules to organize the sentence.
It might be more accurate to say that there’s been a shift from declensional to positional grammar. It’s still a curious phenomenon.
FYI, Latin used to have a separate number for two of an object. The declension survives only in the adjective “ambo”, meaning “both”.
Old English also had dual number for some pronouns. There’s no trace of this left in Modern English.
If you’re really interested in this subject, read this recent book: The Unfolding of Language : An Evolutionary Tour of Mankind’s Greatest Invention by Guy Deutscher. I read it last year, and it does a good job of answering your question-- languages simplify and then become more complicated again in a constant cylce of change.
As Ultrafilter noted, you’re confusing reduction in declension and conjugation with simplification. Languages tend to be synthetic, with nouns’ and adjectives’ sentence function and verbs’ tense, number, and person expressed by suffixes (or rarely by prefixes or infixes); analytic, in which nouns’ positions in the sentence and relationship to prepositions, and verbs’ auxiliaries, carry the same meaning as did the endings. A third group is the agglutinative languages like Turkish, in which a suffix carries a specific single bound-morpheme meaning like plurality, past-tense, possession, etc., and words are constructed of root plus a string of single-purpose suffixes. Juliam ardore amaveram and “I had loved Julia with ardor” have identical meanings, with the significant words carrying identical relationships; only what is used to show person, tense, and grammatical position in the sentence change.
Old English nouns and adjectives, by the way, did in fact have case endings, which were gradually sloughed off during the Middle English period.
There seems to be a tendency for synthetic languages to tend towards analytic-ness in modern times, at least in the Western European language families. Romanian, interestingly, did something different and offbeat. Nouns in and of themselves have no case endings, but when used definitely (“the ball” as opposed to “a ball”) the article, which is declined, is suffixed, making them appear to retain a series of case declensions.
Interestingly, Finnish, from the Finno-Ugric group, is moving in the opposite direction. In Finnish, the “preposition” is actually a post-position, as though we would say “President-with” or “Congress-to” instead of “with the President” or “to the Congress.” And the postpositions are suffixed, giving Finnish nouns the equivalent of one or two dozen cases, though it can be argued that this is actually agglutinative rather than synthetic.
I also came in here to recommend this book. It’s devoted to answering this very question.
A very short summary is that humans tend to be lazy speakers. They drop syllables, elide sounds, and run prefixes, postfixes, auxiliary verbs or other small pieces of language together. Deutscher theorizes that the multiple cases found in Sanskrit and similar languages are really the result of a lost era of simplifications of this sort. However, a further round of simplifications then started wearing away the case endings.
In other parts of languages, however, a series of words get run together (somewhat like agglutination) to form more complex forms. Compound words are built up this way.
The simplification does seem to be related to increased amounts of interaction among non-speakers or at least different speakers of a language. Small, homogeneous tribes with little outside contact tend to keep “complicated” forms because everyone is taught them from birth and no confusion results. More cosmopolitan areas in which not everyone is a native speaker have these complexities worn down to their most basic forms.
He admits that we are not seeing the new cycle of complexification in cases that this theory would imply. No good explanation is available for this yet, but we don’t know what the future will bring in the long term either.
Deutscher covers this in five chapters, one each examing each of the five components of this cycle of change, so no summary can do it justice. It’s good popular science, though, in that he makes the chapters as technical as possible without being overwhelmingly academic. It’s about as good a compromise as I’ve ever seen on language formation. The opening chapter is especially good as an overview.
Yeah, this is not a textbook, but it’s not a breezy read either. There was one chapter (I think it was the one on Semitic languages) that required my absolute full concentration to get thru. And even then, I found myself rereading many sections multiple times before I understood them. That was not the fault of the author (ie, poor writing), but it’s just a tough subject to tackle.
I don’t understand this. Would you mind posting some examples?
It is often said that there are two forces at work in language change: one is the need to be simpler (i.e., easier to say), and the other is the need to be understood. If we made everything really easy to say, then we wouldn’t be able to distinguish between different words and meanings. So we make other things complex. There can also social forces at work in language change, where people want to be more or less alike as some other group that speaks the same language.
I did a quick wikipedia search, and came up with this. I was trying to come up with a Hungarian example for you, but I’m a bit brain-dead at the moment. So the wikipedia site has some examples for ya.
Yeah, I took a look at that. I think I kinda get the idea now, but I’ve got the impression that there’s some reason why German is not agglutinative even with its supernouns, so I’m not 100% just yet.
Exactly. Reducing morphological complexity does not mean that the language has become simplified. It just means that that information is now encoded within the syntax.
A true agglutinative language tacks on affixes for verb tenses, verb aspect, the role of noun and pronouns in a sentence, etc. This is not the same as German “supernouns”, which are essentially strings of adjectives and nouns squashed together to make up what is essentially still a noun in the end.
For somewhat simpler discussions of this topic, see also:
John McWhorter, Tower of Babel: A Natural History of Language
Frederick Bodmer, The Loom of Language
Ultrafilter – you’ve heard of the artificial Klingon language created for Star Trek, right? Linguist Mark Okrand developed that as a textbook agglutinative language. Let me look up some links for you to peruse.
Note that this is different than what’s going on with German “supernouns”. In German, the words denoting the relationship between nouns (prepositions, as in the example above – “for”) are not incorporated into the supernoun – they are still seperate words. Agglutinative languages like Klingon, Hungarian, Turkish, etc. incorporate an affix to serve the purposes prepositions serve in English and German.
[QUOTE=clairobscur]
I’m not sure it’s simpler. By dropping the declensions, you make necessary strict rules to organize the sentence.
For instance, the most difficult idiomatic part of French for me is the prepositions, closely followed by some of the word orders for pronouns. There is no simple equivalence between the English prepositions and the French preprositions. Similarly, if you get the word order wrong in some cases, you may not be properly understood.
In our positional languages, prepositions and word order carry out a key function, equivalent to inflections in inflected languages - so I would not agree that a positional language is automatically simpler. A speaker of a positional language is likely to find it easier to learn another positional language rather than an inflected one, but that is not a measure of the simpliticty of the language. The positional speaker is already primed to understand how positional languages work, and so finds it easier to learn.
Esperanto is good for this: the whole language is a set of little words that you put together like Legos. For example, the word ekriĉiĝinte means “when I have begun to become rich”.
Ek/riĉ/iĝ/int/e:
Ek is a prefix indicating that something is beginning.
riĉ = rich.
iĝ = to become.
int = past active verb tense.
e = adverb ending.
Working out from the root of the word, which is riĉ, we have:
riĉ = rich. It’s a root form; you add endings to make it into a verb or a noun, etc.
riĉiĝ = become-rich. Still a base form.
ekriĉiĝ = start-to-become-rich. Still a base form, but in English you almost have to write it as a verb.
ekriĉiĝint = have started to become rich. Still a base form.
ekriĉiĝinte= have started to become rich. Now it’s an adverb, and you can use it in a sentence.
I disagree that there is in language a “need to be simpler”. Language should be regarded as a natural human behavioral trait, much like walking, or even an inanimate component of the physical world, such as gravity. It doesn’t “need” anything.
We still need to know why some languages use inflection, and others skew toward syntax. One explanation I’ve heard is that languages whose speakers suffer invasion and conquest by speakers of a foreign language tend to become morphologically simpler. Parts of England had already undergone domination by the Danes, and although their language at the time was somewhat similar, confusion beetween the two was already leading to inflection loss prior to the time of the Norman conquest. The Norman Conquest was like an atom bomb with regard to Old English, as that langage was practically eliminated from aristocratic and educated circles. There are few such complete conquests in history; for example, the Germanic invaders of Italy, France, and Spain 600 years earlier tended to adopt the Romance speech of their subject lands, and rely on Roman civil officials for purposes of administration.
Contrast the condition of English against that of Icelandic, whose speakers were never conquered, and who have been joined by very few immigrants. Their language is as complex as ever.
I mean, the inflection system is as complex as ever.