Why is Russian so Complicated?

[csawyer]

Why is Russian so complicated?

German is almost an analytic language in comparison to Russian -- cases pared down to 4, no inflections except for adjectives and articles, no aspect, all the non-indicative moods either abolished or re-labelled as subjunctive (Subjunktiv II of course is not subjunctive at all), no trace of the dual number.

Why does Russian have all of this and German not?

Russian even has some things which are gone in Classical Latin (aspect fully expressed in all the tenses, dual, etc.). Why? Were the Slavs so literate that they preserved the old grammatical structures while the Germans did not? Or was it because Old Church Slavonic was close enough to the vernacular that the vernacular was constantly replenished? Whereas the Germans only had Latin as a literary language, which could not be much used to replenish German as Germans forgot their own grammar?

Another thing that bothers me about Russian -- the alternation between nominative and genitive endings for the accusative, when dealing with animate nouns. Is Russian partially or vestigially ergative? Note that Finnish is even more so like this -- there isn't even a real accusative case (although they have 15 of them). Finnish and Russian are unrelated, but the Russians lived cheek to jowl with Finno-Ugric peoples for a thousand years or more -- could this have worn off on the Russians? Or what?

 

[ahvalj]

Another thing that bothers me about Russian -- the alternation between nominative and genitive endings for the accusative, when dealing with animate nouns. Is Russian partially or vestigially ergative? Note that Finnish is even more so like this -- there isn't even a real accusative case (although they have 15 of them). Finnish and Russian are unrelated, but the Russians lived cheek to jowl with Finno-Ugric peoples for a thousand years or more -- could this have worn off on the Russians? Or what?

That's more or less a common Slavic trait. Leaving aside the (numerous) details, the nominative and accusative coincided in the singular of the most widespread declension type in most Slavic dialects as a result of certain phonetic changes in the last centuries of the 1st millennium, and, since at least in animate nouns the distinction between subject and object was considered important, speakers began to use the genitive to denote the object, following the pattern found in negative sentences:
вижу сестру — не вижу сестры
вижу брат — не вижу брата → вижу брата.
This use of the genitive instead of the accusative after negation occurs in Slavic (more in older texts, less so in the modern times), Baltic and Baltic-Finnic, so, while the use of genitive for the accusative in animated nominals is not related to the Baltic-Finnic usage (as only future Russians came into contact with Finnic speakers), the genitive after negation may be a common areal feature in northern East Europe, though a considerably older one.

Along with Slavic, the genitive is used instead of the accusative in Ossetic, a language descended from the speech of Sarmatians that were in contact with Slavs in antiquity, but here it is only used with personal names and not after negation, so it unclear if it may represent a shared innovation or is just a coincidence.

 

[csawyer]

Wow, thanks, very informative.

So what you are saying is that this is not about some ambiguity between the cases, but rather, the inflections for accusative and nominative simply coincide, and so a distinction was re-introduced (by whom?) in order to distinguish between accusative and nominative, and the genitive was grabbed for this purpose.

Did I understand you correctly?

And do we know this is really true? Considering the fact that there really is ambiguity between these cases in some languages, at least in the ergative languages accusative is not what it is for us.

 

[ahvalj]

Yes, of course, compare the singular of the а-declension where there is no merger of the nominative and accusative and hence no use of genitive (вижу маму/папу), or the singular of the ь-declension, where, despite the merger, the genitive didn't penetrate into the accusative (вижу мышь).

In the plural, the pervasive use of the genitive for the accusative is specifically modern East Slavic; e. g. in Serbo-Croatian the nominative plural is -i, the genitive is -a and the accusative is -e. If we look at the Slavic texts of the turn of the 1st and 2nd millennia, the use of genitive for the accusative only characterizes the singular of the o-declension, and even there it is inconsistent.

 

[csawyer]

Brilliant explanation; thank you. Now I think I understand for the first time something my Russian teachers were never able to explain. I owe you a beer.

Now how about the first part of my original question?

 

[ahvalj]

I think what has happened to Germanic languages is first of all a tragic outcome of the reduction of the (unstressed) endings and the trend towards morphological simplification it set. This, however, is only part of the answer as e. g. Italian didn't experience any significant phonetic reduction (though it lost the Latin -s and -m, which were very important for the morphology), but lost the declension while at the same time preserving (and incrementing) the complexity of the verbal system. As an opposite example, Lithuanian, the most conservative living Indo-European language, has mostly retained the endings and the rich and complicated nominal declension, but has greatly simplified its verb (comparing with Greek and Sanskrit), while creating new, mostly analytic, verbal forms. So, the honest answer would be that the language evolution is caused by phonetic changes + some internal logic + plain fashion.

 

[OBrasilo]

The Slavic languages have Bulgarian and Macedonian who evolved much the same way as the Romance languages - they largely lost their case system and replaced it with prepositions, but developed (enclitic) articles, and not only retained the old Slavic verbal system, but made it more complex.

 

[danielstan]

Well, the loss of case system in Bulgarian and Macedonian is considered the biggest influence that Proto-Romanian (the Romance language spoken in Balkan Peninsula before Slavic invasion) had on Middle Bulgarian.
There was a Romanian - Slavic symbiosis in this area,
which resulted in Slavs assimilating the Vlachs in the territory South of Danube and Romanians assimilating the Slavs at North of Danube.
Modern Romanian has in its core vocabulary Slavic terms related to family members, love, friendship and other social relations; also Romanian has Slavic calques - all suggesting inter-ethnic marriages and Romanian - Bulgarian bilingualism that happened during Middle Ages.

 

[Olaszinhok]

Why is Russian so complicated?

Does Russian have articles (definite, indefinite, partitive, no article, etc.), compound prepositions, the subjuctive mood and many verbal tenses and moods? These are just a few grammar features other Languages have and Russian doesn't...
Isn't English syntax more unpredictable (and definitely complicated) than the Russian one?

 

[francisgranada]

Ciao Olaszinhok. I fully agree with your comment (post #9).

 

[ahvalj]

I think the original question represents one particular comparison, whereas a more general issue is that (1) languages don't enrich with time, they always get and lose, and sometimes lose dramatically, and (2) unlike in many other areas of human activity, languages are not equivalent in their expressive power, even not basically equivalent, and one language will always be more nuanced than any other one in some moments and more restricted in the others.

 

[Hulalessar]

Were the Slavs so literate that they preserved the old grammatical structures while the Germans did not?

Literacy has no relation to grammatical features perceived to be "old", by which, from the rest of your post, I assume you mean something like "bristling with inflections". Many unwritten languages are more synthetic than Russian. In fact, even languages such as Latin, (Ancient) Greek and Sanskrit are only moderately synthetic compared to others.

 

[thegreathoo]

(the Romance language spoken in Balkan Peninsula before Slavic invasion)

There was no slavic invasion. There was however, Roman invasion, well documented.

 

[Awwal12]

Why does Russian have all of this and German not?

Mostly because proto-Slavic had it and proto-Germanic already didn't, it seems.

Why is Russian so complicated?

Does Russian have articles (definite, indefinite, partitive, no article, etc.), compound prepositions, the subjuctive mood and many verbal tenses and moods? These are just a few grammar features other Languages have and Russian doesn't...
Isn't English syntax more unpredictable (and definitely complicated) than the Russian one?

While the English system of articles is frankly mind-boggling, Russian mostly makes up for that with its complex communicative strategy (an inherent part of any language which is often left forgotten somehow), which results in complicated word order, intonational patterns and rules of using discourse markers (English is nowhere close in that regard). In the aspect of language learning it's even worse than any articles, because communucative features aren't related to formal semantics at all and their usage is particularly difficult even to correctly formulate (to the note, scholars still widely use conflicting terminologies and descriptions in that field).

Russian (as well as all Slavic languages) has a very simplistic system of moods, that's really true.

The tense system seems very simple, but the catch is that it cannot be considered independently of the aspectual system, and the latter is very bad news (since aspects are essentially a LEXICAL feature, the resulting system ends up full of all kinds of irregularities - as if it weren't complex enough even without that; remember those SUDDEN usages of the aspects in a good half of different negations).

"Syntax" is too broad a term. Constructions with dependent clauses (finite and non-finite) are a bit more complex and irregular in English, but it doesn't make a really substantial difference. What else you might have in mind?..

Don't forget the Russian numerals (which are a headache for both native speakers and foreign learners, although for different reasons).

Overall Russian seems more complex than English grammatically - meaning it has more complex rules and more exceptions (you can safely call the system of declension one big exception: officially there are "just three declension paradigms", but if we take into account all the nuances, - the stress patterns in particular, - the number of actual paradigms may reach a hundred). But you're right that in the eyes of an English speaker that comparative complexity will be usually overrated because he'll naturally ignore all the areas where Russian is actually simpler.

 

[jasio]

Perhaps, considering the history of Europe, the question should be: why the languages of the western Europe were mixing towards creolisation, while the slavic languages in general did not - with a few notable exceptions like the said Bulgarian and Macedonian.

We tend to perceive romance languages as descendants of Latin, but in fact they are mixtures of a local dialect of vulgar Latin with Celtic, germanic, Greek, Arabic, slavic, and perhaps other admixtures in various proportions. Some of them being substrates, while the others only influence vocabulary and pronunciation. Only Icelandic was pretty much isolated, and to the lesser degree, Nordic languages.

BTW, as far as I am aware, natural languages in general do not develop just to upset foreign learners, and for the native speakers the complexity here and there is less relevant than providing an effective communication tool.

 

[Olaszinhok]

Europe were mixing towards creolisation, while the slavic languages in general did not - with a few notable exceptions like the said Bulgarian and Macedonian.

Sorry, that is utter nonsense from a linguistic point of view! You statement does not make any sense.

 

[Zec]

Sorry, that is utter nonsense from a linguistic point of view! You statement does not make any sense.

I agree, but with a caveat. I certainly thing it's wrong to ascribe every case of apparent language simplification to "creolisation". Losing some or all cases is not creolisation, creolisation is, somewhat jokingly, destroying the entire grammar and rebuilding it from the ground up! Anyone who has taken a glance at Tok Pisin, for example, will be able to see that it's grammar is entirely different from English and even much of vocabulary is used in unexpected ways. Calling Bulgarian/Macedonian a creole for losing cases when it has an extremely nuanced verbal system where the aorist, the imperfect and the perfect tenses can all be used in perfective and imperfective aspects is simply wrong.

However, there does seem to be a kind of societal influence on language, at least I've read so. Supposedly large, widely spoken languages tend towards simplification, while small, isolated languages tend to accumulate complexity over time, possibly because in small communities the not strictly communicative aspects of language get elaborated upon.

 

[Olaszinhok]

Calling Bulgarian/Macedonian a creole for losing cases when it has an extremely nuanced verbal system where the aorist, the imperfect and the perfect tenses can all be used in perfective and imperfective aspects is simply wrong.

We can certainly say the same for the Romance languages: most of them have a very inflected and nuanced verbal system. Anyway, I wouldn't call that "simplification". For instance, Russian and Polish have inflected nouns and adjectives but the use of articles, prepositions, verb tenses and syntax is generally much more nuanced in English, for instance. There is some sort of compansation, in my opinion.
Portuguese, spoken in the far west of mainland Europe, boasts a pretty complicated verbal system with loads of tenses and moods and even a conjugated infinitive.
Besides, Nordic languages, apart from Icelandic and Faroese, are very analytical languages, much more than the Romance languages...

 

[ahvalj]

And, since it has been mentioned twice here, what's so particularly nuanced in the English syntax comparing with Russian (or Polish), other than the presence of gerund (which also exists in Polish, while Russian retains participial constructions, much restricted in both English and Polish)? Translators of scientific texts from Russian into English always struggle with repackaging long complex and loose Russian sentences into their English, much shorter and very straightforward, counterparts. (This English approach is beneficial to clarity, as Russian writers often fail to express clearly what they mean, but the syntax of an average Russian page is definitely not more restricted than in English).

 

[Awwal12]

English uses infinitive phrases much more willingly and in a more complex fashion, for one thing.

Translators of scientific texts from Russian into English always struggle with repackaging long complex and loose Russian sentences into their English, much shorter and very straightforward, counterparts.

"Shorter" doesn't mean "fundamentally more primitive". It only indicates one thing: Russian scientific language (especially in scientific papers) tends to have overcomplicated structure (usually the more useless the paper is the more complex language it utilizes). As far as I recall, a French legal sentence is traditionally formulated, indeed, in one sentence (no matter how many pages long). We can only thank God that at least there is no such requirement on a Russian scientific paper: while formally possible, it would make it positively impossible to read.

 

[ahvalj]

Infinitive constructions (I saw him do something) are indeed useful, but these are just a nice way to convert a compound sentence into a simple one, they make the sentence lighter but they don't express additional nuanced meanings.

Update. Though I myself meant elegance in #19. OK, English has infinitival and (especially) gerundial constructions, while Russian has participial and adverbial participial constructions, 1: 1. What else?

 

[Zec]

I can't speak for other Slavlangs, but at least my native Croatian packs a lot of modality into conjunctions instead of into the verb (I noticed that some conjunctions differ from others chiefly in expressing different modal values, such as mirativity for example). Our coordinating conjuctions are especially nuanced.

 

[Awwal12]

OK, English has infinitival and (especially) gerundial constructions, while Russian has participial and adverbial participial constructions, 1: 1. What else?

Such a straightforward comparison hardly can tell a lot. The more interesting question is how exactly the languages use their parts of speech to produce more complex structures, how complex are those patterns themselves and how many exceptions they contain.

(By the way, English gerunds are an extremely powerful tool, even though English speakers very rarely use them to their full capacity.)

 

[ahvalj]

In this case the syntax of the Turkic type (with few if any compound sentences and a rich system of verbal nouns) has no competition.

 

[Hulalessar]

I agree, but with a caveat. I certainly thing it's wrong to ascribe every case of apparent language simplification to "creolisation". Losing some or all cases is not creolisation, creolisation is, somewhat jokingly, destroying the entire grammar and rebuilding it from the ground up! Anyone who has taken a glance at Tok Pisin, for example, will be able to see that it's grammar is entirely different from English and even much of vocabulary is used in unexpected ways. Calling Bulgarian/Macedonian a creole for losing cases when it has an extremely nuanced verbal system where the aorist, the imperfect and the perfect tenses can all be used in perfective and imperfective aspects is simply wrong.

However, there does seem to be a kind of societal influence on language, at least I've read so. Supposedly large, widely spoken languages tend towards simplification, while small, isolated languages tend to accumulate complexity over time, possibly because in small communities the not strictly communicative aspects of language get elaborated upon.

Rather than asserting that any of what Jasio said is nonsense, I would say that what he means depends on what he understands by "creolisation". We have discussed that question elsewhere: Is English a creole?. My view is that if creolisation is to be a useful term it refers exactly to situations a new language arises phoenix-like from the ashes of two or more different languages.

Languages meet and mix in many different ways and proportions. You can start by asking whether any given language has one parent or two. Broadly speaking, most languages have only one parent. A simple situation is where variety A splits into varieties A1, A2, A3...An and develop without any contact with each other. In practice what happens where the speakers of the varieties remain in contact with each other is that they influence each other. The influence may be asymmetrical depending on all sorts of different circumstances. The position gets more complex when any of the varieties comes into contact with or is adopted by speakers of a variety which is not A. Contact may involve any of the various "strates" in different degrees, but without involving the extreme of a pidgin which develops into a creole.

Each of the Romance languages may have different strates according to the languages spoken before Latin arrived and later influences, but they all follow the same ground plan and have more in common than divides them. If you suggest that, say, French is a creole, you are well on the road to suggesting that a huge number of languages are creoles.

The models of language history are the tree model and and the wave model. Both are useful. The tree model shows the broad genealogy and is easy to follow, while the wave model shows the interactions and can be confusing and is in practice necessarily incomplete as it cannot show everything. You just have to accept that languages, like much else, cannot be neatly packaged.

As to complexity, you can start by accepting that there are basic things that all humans experience. One is that things do things and things have things done to them. All languages can tell us if the dog is chasing the cat or the cat is chasing the dog. Doing it by marking the nouns for case is not inherently more complex than doing it by word order. Children soon grasp how to it however a language does it.

As for the supposed complexity of languages used by small communities, you have to be careful to distinguish between the actual language and baroque formulae consciously introduced and used only in limited situations. Many widely spoken languages are not without features which not everyone may master and which may be required in certain social situations.

 

[tunaafi]

I have, in my time attempted to learn French, German, Czech, Russian, Mandarin,Turkish and Latin., with varying degrees of success (perhaps I should say failure).

I have certainly found some more difficult than others, but, as a linguist by profession, I can't say I have found any of them more complicated than any of the others. The things that learners often find complicated about another language are features that don't exist in their own language. It is still almost always possible to express a thought that uses on feature of grammar in one language by a different feature in another.

 

[Olaszinhok]

I have certainly found some more difficult than others, but, as a linguist by profession, I can't say I have found any of them more complicated than any of the others. The things that learners often find complicated about another language are features that don't exist in their own language. It is still almost always possible to express a thought that uses on feature of grammar in one language by a different feature in another.

I couldn't agree more...

 

[francisgranada]

I couldn't agree more...

So am I ....

 

[Red Arrow]

However, there does seem to be a kind of societal influence on language, at least I've read so. Supposedly large, widely spoken languages tend towards simplification, while small, isolated languages tend to accumulate complexity over time, possibly because in small communities the not strictly communicative aspects of language get elaborated upon.

Then why did Russian and German keep their case systems while the smaller languages Bulgarian and Dutch lost theirs?

Perhaps it makes more sense to conclude that isolated languages are more conservative. Iceland is more isolated than Norway. Wherever in Russia Standard Russian comes from, it might have been more isolated than Bulgaria and Macedonia, which share a border with Turkey, Romania, Albania, Greece etc.

Change does not always make things easier. Estonian has got 44 nominal classes because vowels got dropped and words got shorter, while Finnish and Hungarian follow conservative (but regular) rules.

 

[Awwal12]

Perhaps it makes more sense to conclude that isolated languages are more conservative

...And you instantly made a reversed assumption that the more conservative is the language the more isolated it must have been. Though even if the original assumption is valid, it doesn't automatically validate the second one.

Change does not always make things easier.

True. Say, Old Russian had more complex grammatical rules - but it also was much more regular than Russian is.

 

[Red Arrow]

...And you instantly made a reversed assumption that the more conservative is the language the more isolated it must have been. Though even if the original assumption is valid, it doesn't automatically validate the second one.

My "assumption" (more like a hypothesis?) is not derived from Zec's one. I simply read it somewhere else. (Marc van Oostendorp)

 

[Awwal12]

I'd re-fotmulate it entirely, merely stating that intensive linguistic contacts tend to intensify linguistic development (having creolization as the upper limit).

As for Bulgaro-Macedonian dialects, one shouldn't concentrate too much on the loss of noun cases (which, by the way, didn't occur in all the dialects), because, for example, their verbal systems are most archaic and complicated among the living Slavic languages. Russian, in turn, is archaic in certain respects, but demonstrates intensive development in other areas (take the vowel system, for instance), sometimes shared by other mainstream Slavic languages and sometimes unique.

 

[Zec]

Re: small languages being complex. The context of the statement was an attempt to explain why languages spoken by small tribes of hunter-gatherers often have crazy complexities and irregularities. I agree with Awwal12 that it'd be safer to state that linguistic contact tends to intensify linguistic development. In the context of Bulgarian and Macedonian it would be whatever circumstances led to the development of Balkan Sprachbund and these two languages developing in a different direction than other Slavic languages.

 

[ahvalj]

I think it's not the matter of complexity, only of relative conservativeness. Complexity may represent a survival of some older stage or a novelty arisen from phonetic changes: Old French was not a minor language yet it had very complicated morphophonemics with various complex alternations within paradigms (mostly leveled out in the modern times, but cp. veux — voulons).

 

[jasio]

I see that an interesting discussion arose from my comment - although I did not follow it for a while. It seems however that my point was completely misunderstood, so let me try to clarify it a bit.

Perhaps, considering the history of Europe, the question should be: why the languages of the western Europe were mixing towards creolisation, while the slavic languages in general did not - with a few notable exceptions like the said Bulgarian and Macedonian.

Inspired by Is English a creole? discussion perhaps I used the word 'creolisation' a bit too lightheartedly, with the sole purpose of enhancing the difference of evolutionary trajectories of the Slavic languages vs. Romance and Germanic languages. I must have overdone though, if my main point was generally missed, and the discussion focused on whether 'creolisation' as such was used properly.

The issue is, albeit it's convenient for clasification purposes, the tree model does not fully represent the history of the European languages. For example, standard italian - albeit often perceived as a pure descendant of Latin - was a product of mixing local vulgar Latin with the germanic language of the Longobards (not mentioning other influences), on an earlier local Celtic stratum. The French is a product of a local vulgar latin on a local celtic stratum mixed with another germanic language of the Franks. So are English and Spanish. In all these cases the grammar complexity of the Latin was almost entirely got rid of, and their grammars were to a large extent constructed anew using other linguistic tools, such as an extensive use of prepositions, more rigid phrase structure, articles, complex tense systems, etc. What was left of the original Latin was chiefly vocabulary.

On the other hand, Slavic languages mostly retained their complex case and aspect system, relatively simple tense system, yet imported primarily the vocabulary adapting it to existing inflection patterns. This is the case with Polish at least, although we have a 1000 years' history of close contacts with German (most cities and some rural areas were predominantly German-speaking until some 18th century), Latin (which was an official language of the state the Church, and was broadly spoken by the gentry and educated people), and French, not mentioning Turkish and neighboring Slavic languages including Ruthenian (AKA Old Russian) and Czech.

The same goes for Russian. Someone suggested that Russian had only limited contacts with other languages - but this is not the case. A mere fact that on modern maps Russian is shown as a big, red spot does not mean that other languages are not or were not used in the same territory. In fact, Russian and Old Russian have quite extensive contacts with UgroFinnish, Baltic, Turkic, Greek, and many more languages which had been used on the lands which are currently associated with Russia. Not mentioning other Slavic languages, including Old Church Slavonic and Polish.

So my question is: why the West-European languages generally simplified their case systems and some other aspects of their grammars, while the Slavic languages in general did not.

 

[jasio]

Separately, albeit I agree that comparing complexity of the languages is not a football game, I would not agree that the structure of analytic languages has the same complexity as synthetic languages. Even if the children learn respective structures at similar paces, I would rather claim that synthetic languages have higher entry barriers than analytic languages, at least for foreigners. Let's take the simple phrase which was quoted earlier in the discussion:

the dog chases a cat

That's pretty straigtforward and even if you have only limited grammar understanding and say

*dog chase cat

you still end up with an understandable statement, isn't it? In Spanish it functions quite alike:

el perro persigue a un gatto

*perro perseguir gatto

would require a bit of thinking to decode unusual structure, but would still mean the same.
In both of these languages you can communicate at the basic level quite effectively using just the SVO structure and a handful of words.

But what about:

kota (Acc.) goni pies (Nom.)

If you simply disregard the inflection and say

*kot gonić pies

*cat chase dog

you end up with a phrase which is not only incorrect, but also has quite an opposite meaning than the intended. And in case of such a simple statement in Polish (and in most other Slavic languages) you can put the words in any of the six possible orders without any change of the meaning whatsoever - perhaps just the nuances. So to understand or express even the basic messages you must be able to use at least two-three cases in most popular conjugations (and recognise more of them). Actually, there's even a children's rhyme based on this very phenomenon:

A było to tak:
Bociana dziobał szpak.
A potem była zmiana,
I szpak dziobał bociana
Później były jeszcze trzy takie zmiany.
Ile razy szpak był dziobany?

 

[Red Arrow]

This is the case with Polish at least, although we have a 1000 years' history of close contacts with German (most cities and some rural areas were predominantly German-speaking until some 18th century), Latin (which was an official language of the state the Church, and was broadly spoken by the gentry and educated people), and French, not mentioning Turkish and neighboring Slavic languages including Ruthenian (AKA Old Russian) and Czech.

In the Germanic languages, the case endings are all unstressed. Unstressed vowels turn into schwas, schwas are dropped, as are often the consonants R, N and T. (think of Dutch or English) This leads to case endings being dropped.

In Polish, the stress is always on the last syllable. I don't know much about Polish, but to me it looks like it has also dropped vowels (like in Źdźbło), but no case endings because those are stressed.

 

[jasio]

In the Germanic languages, the case endings are all unstressed. Unstressed vowels turn into schwas, schwas are dropped, as are often the consonants R, N and T. (think of Dutch or English) This leads to case endings being dropped.

In Polish, the stress is always on the last syllable. I don't know much about Polish, but to me it looks like it has also dropped vowels (like in Źdźbło), but no case endings because those are stressed.

I get your point, and you may be right with the Germanic languages, but for the Slavic languages it does not work that well.

In Polish the penultimate sylable is accented, not the ultimate. Although depending on the length of the suffix it could indeed be accented, the fixed accent can protect the cases to some extent because dropping a syllable would require changing the speach pattern - and this is quite unlikely. The Polish accent is quite stable and nowadays I observe that more and more words, which originally were accented differently and deviated from the regular rule, are nowadays accented in the regular way (even if it's perceived incorrect by the linguists*). And it worked quite similarly in the past: during disappearance and vocalization of yers the syllable structure was reconstructed to preserve the rhythm of the spoken language, which eventually produced a lot of complications for modern learners (pies, psa, psu, psa, psem, psie). So the regular accent in Polish could have contributed to the preservation of the cases even if the suffix itself was not accented.

But what about the languages like Russian, with its entirely free and movable accent? Or the Czech and Slovak languages (and some Southern dialects in Poland), in which always the initial syllable is accented, so the suffixes are never accented?



*) the words of Greek origin, like fizyka, matematyka, gramatyka are now often pronounced as fizyka, matematyka, gramatyka. Also some verb forms in which the former auxiliary verbs or particles had been assimilated as suffixes, and for centuries were accented according to the original patterns: byliśmy, bylibyśmy (from byli śmy, byli by śmy, standing for "we were" and "we would be") are now more often pronounced as byliśmy, bylibyśmy.

 

[Red Arrow]

Oops, sorry, you are correct. Polish stresses on the penultimate syllable just like is often the case in Spanish... but Spanish lost its grammar cases and Polish didn't. I don't know why that is.

 

[Hulalessar]

When comparing French and Spanish with Latin it can be said that the morphology of nouns has been simplified. However, when it comes to verbs, whilst the number of forms has been reduced, complexity has increased because of the number of the number of irregularities which have been appeared. There are only about a dozen irregular verbs in Latin. My reference books have 83 different tables for French and 65 for Spanish. Some of the irregularities are purely orthographic, but they still have to be learned. Having studied Latin, French and Spanish at school I would say that it takes more effort to master the forms of French and Spanish verbs than Latin verbs.

We have had a few discussions on complexity. Complexity tends to be equated with complex noun and/or verbal morphology. When a language has them you have to get to grips with them early on otherwise you are not going to make progress. The complexities of English are of a different order but may not be perceived as such since they do not involve "Amo, amas, amat" type rote learning. It is also difficult to see how complexity can be measured. How do you compare the devilshness of Spanish radical changing verbs with the opaqueness of English phrasal verbs?

 

[jasio]

However, when it comes to verbs, whilst the number of forms has been reduced, complexity has increased because of the number of the number of irregularities which have been appeared.

Aren't "irregularities" as such byproducts of the descriptive grammar? I've heard somewhere that when Panini encountered phenomenons which did not fit within existing rules, instead of marking them 'irregular', he created a new rule. And indeed, many 'irregular' verbs exhibit traces of older grammar rules, and are in a sense similar.

The complexities of English are of a different order but may not be perceived as such since they do not involve "Amo, amas, amat" type rote learning.

Geez, I spent weeks and months walking 'round my house and repeating 'have, had, had', 'put, put, put', 'get, got, got', 'steal, stole, stolen', 'held, hold, hold' - and now you're telling me that I was only wasting my precious time? ;-)

It is also difficult to see how complexity can be measured. How do you compare the devilshness of Spanish radical changing verbs with the opaqueness of English phrasal verbs?

...or with the Polish verb conjugations which are based on the known conjugated forms of the verbs rather than on the infinitives like -ar(e)/-er(e)/-ir(e) in Romance languages. ;-)

It's not simple indeed. And the additional question is, what level of the language comprehension and what registers are we going to compare: whether an A1/A2 level beginner, precise yet complex legalese or a rhyming slang in which a the words rarely mean what one could expect.

A number of existing word forms, a number declension patterns, and a number of 'irregular' words - whatever it would mean in the shed of my remark above - could give a hint though. You can also compare similar phenomenons one by one to see if they produce any meaningful patterns between the languages.

For example Slavic gender/category system, which embraces differentiation between inanimate, animate and personal noun classes (at least in some languages) and influences both verbs and adjectives, is objectively more complex than two grammatical genders (even, if you consider exceptions, such es 'el agua') in Romance languages (italian and spanish, at least) or no gender at all, as in English. So are adjectival declinations, where the adjective not only has to agree with the gender and number, but also has to assume the correct grammatical case to match the noun. Unlike again - Spanish or Italian where the adjectives only have singular/plural and masculine/feminine forms. Or English, where "red" is always "red" regardless of the grammatical context. On the other hand, the tenses in English - even if some of them do exist only in theory - Italian or Spanish are objectively more complex than the Slavic past-present-future (ok... in practice it's a bit more complex because of the verb aspect and aorist in some languages, but still I do not think we could compare to eight tenses in italian, even disregarding conjuntivo).

A side note: this reminded me a discussion somewhere, when a Dutch lady asked me 'if you do not have articles, how do you distinguish between the noun cases?'. Whatever complications are related to the declension of articles, it's a peanut compared to the declension of nouns, believe me. ;-)

 

[Hulalessar]

Aren't "irregularities" as such byproducts of the descriptive grammar? Etc

When writing my last post I put "irregularities" in quotes and then changed my mind. I discussed the use of the word in post 135 on this page Complex morphological structure of IE languages in comparison to other language families. From a historical perspective some irregularities are no more than instances of sound changes that occur throughout a language and in that sense are perfectly regular. “Irregular” is though a useful word when teaching and in a thread like this where it means something like “follows a paradigm that lots of other words do”. It may be arbitrary, but there is at least some method to it and it is also useful. In a discussion about complexity whether you say Spanish has three conjugations and lots of irregular verbs or 65 conjugations it comes down to the same thing: there is a lot of spadework to be done. Anway, my point is that to say that Spanish verbs are less complex than Latin verbs homes in on the reduction of the number of forms and ignores the increase in variation.

Consider the verb “to ask”. In Latin the present tense go like this:

rogo
rogas
rogat
rogamus
rogatis
rogant

In Spanish it goes like this:

ruego
ruegas
ruega
rogamos
rogáis
ruegan

The Spanish forms reflect sound changes affecting stressed syllables which occur across the language. In that sense they are regular, but they nevertheless introduce a complication not found in Latin. Rogar is just one instance of many where Spanish introduces complications not found in Latin. My subjective opinion is that, solely considering the forms, the Spanish verbal system is at least as complex as Latin’s, if not more so. I certainly found it more confusing when I was a boy.

Your anecdote about the Dutch lady highlights that the average person regards anything which deviates from their own language as a complication, or at least an oddity. This thread has concentrated on complex morphology, but is is worth pointing out that languages which are highly analytic or isolating are often found tricky to get a hold of by speakers of synthetic languages because they feel they lack precision.

Someone somewhere said that there are languages where you have to know everything before you can say anything. That may be an exaggeration, but there is some truth in it and it applies equally to languages with complex and simple morphology.

*See also this thread: https://forum.wordreference.com/thr...nguage-isnt-that-difficult-after-all.2746680/

 

[Awwal12]

From a historical perspective some irregularities are no more than instances of sound changes that occur throughout a language and in that sense are perfectly regular.

If they are regular in the etymological sense, it doesn't necessarily make them regular from the point of modern morphology, since the original phonetic differences which triggered the changes may be long lost. Obviously no one learns Russian by studying its development from the proto-Slavic stage.

 

[Hulalessar]

If they are regular in the etymological sense, it doesn't necessarily make them regular from the point of modern morphology, since the original phonetic differences which triggered the changes may be long lost. Obviously no one learns Russian by studying its development from the proto-Slavic stage.

I agree. My last post was responding to jasio who was suggesting that irregularity was something invented by grammarians. From one perspective that is a valid observation. However, my point is that if considering complexity irregularity has to be taken into account.

 

[jasio]

I agree. My last post was responding to jasio who was suggesting that irregularity was something invented by grammarians. From one perspective that is a valid observation. However, my point is that if considering complexity irregularity has to be taken into account.

It actually very much depends on what you actually mean as "irregularity".

Probably in every given language you may encounter the words which do not fit some basic rules. The issue is, what you do about them. You may say "they are irregular words, here's the list for you to memorize" or you may say "there is an extra rule for these words: if the word belongs to a class xyz, which can be recognised by abc, def features, then it behaves like that (for example, there is an o -> ue ablaut)". Using one approach you end up with a long list of isolated (irregular) words tomemorize, while using the other - you end up with a long list of rules which may or may not be easy to use in practice, but with only a limited number or no exceptions at all. In this sense, irregularities are not part of the language, but of the description of the language, ie. they were invented.

Anyway, for a reason I do not have sleeples nights because of the aformentioned Spanish o -> ue ablaut. There is an L rule for them, now I understand that it's a matter of an accented o (thank you) which also appears in many quite regular words which I know with the original "o" from other languages (like porto -> puerto, corpo -> cuerpo, etc.), so it should be easier for me to spot them now.

Now imagine, that you could have such irregularities not only in verbs (with their 6 forms in the present tense), but with all the words, including 14 (2 * 7) forms of nouns, even more forms of adjectives (they have to match the grammatical case and the gender of the noun) and numerals. Someone counted that in Polish there are as many as 17 different forms of the numeral "two", while in English there are only two of them "two" and "second" (or "dos" and "segundo" in Spanish - although I'm not 100% sure if this list is complete). In English there are two participles, past and present, the Spanish has three (present - which perhaps may not be considered a participle any more, gerundio and past), while Polish has four (not mentioning the past participle, which for a strange reason is not considered a participle in the Polish grammar), and Russian has six, if I counted correctly (not mentioning the original past participle, which now is a past tense form for all persons).

That's why I still claim that the grammar of the Western languages is much more simplified than of the Slavic languages.

 

[Awwal12]

and Russian has six

It depends on the verb in the first place. Perfective verbs have past active and past passive participles (no future forms in modern Standard Russian, past participles of perfective verbs do double duty).
Imperfective verbs may have 4 proper participle forms (past and present, active and passive), although passive forms are often formed from the reflexive counterparts, - very often as the only option, - and certain verbs simply have lacunas in the word formation paradigm (typically using the present participle as the past one as well), so the main problem is not that there are up to 4 proper participles, but that you hardly ever know what they will look like, having simply to learn the set every time.
Generally it looks as follows (for masculine singular nominative forms ):
делающий - делаемый (?)/делающийся - делавший - деланный (?)/делавшийся
сделавший - сделанный
любящий - любимый - любивший - любимый
бьющий - (0) - бивший - битый
etc.

Of course, there are also adverbial participles, and in that case there are no morphological means to form passive ones directly, so the reflexive adverbial participles become the only possible variant for passive meanings - although not a guaranteed one.

So, ultimately everything comes to the manner of counting participles - and their very formation is essentially irregular.

 

[Hulalessar]

It actually very much depends on what you actually mean as "irregularity".

Probably etc

"Irregular" is recognised by dictionaries as a term used in grammar. e.g. Departing from the usual pattern of inflection, derivation, or word formation. Some do not like the term because they feel that it implies a value judgement. They may be thinking of situations such as where a Spanish speaker says: "Russian is difficult because it has a lot of irregularities." That prompts the reply: "Excuse me, but so does Spanish." The point would be made that the Spanish speaker may not recognise the irregularities in Spanish. Having a term to distinguish between forms which follow a standard pattern and those which do not is useful when describing a language and particularly when teaching it. We could use a word like "exceptional", but personally I see no need. Rejecting established terms tends to lead to confusion.

In any particular language you have to decide what you are going to class as irregular. In French we can I think distinguish four classes of verbs.:

1. Verbs which follow a rule such as: take -er off the end of the infinitive and add -e, -es, -e, -ons, -ez, -ent for persons 1,2,3 s and 1,2,3 p respectively.
2. Verbs which follow 1 but require a change to comply with the rules of French orthography, e.g. manger which requires an e before -ons and commencer which needs a cedilla unde the c before -ons. (If French orthography was totally phonemic this class would not exist.)
3. Verbs which follow 1 but need a change in orthography to reflect a change in the pronunciation of the vowel in the stem, e.g. lever where you have je lève but nous levons and jeter where you have je jette but nous jetons.
4. All other verbs, that is those which have a unique paradigm or a paradigm shared by a very small number of verbs.

A French speaker familiar with the rules of French orthography may not regard verbs in class 2 as irregular on the basis that he knows the e and cedilla are needed. He may tend to think of those in class 3 as being on a par with those in class 2. He is likely to concede that those in class 4 are irregular. We could call class 1 "regular", class two "orthographic changing", class 3 "slightly irregular" and class 4 "irregular". We could also decide not to call them anything. Whatever we do you cannot avoid the fact that a learner needs to get to grips with a complex system.

Now imagine etc

That's why I still claim that the grammar of the Western languages is much more simplified than of the Slavic languages.

That assumes that complexity involves only morphology and ignores syntax, not to mention phonology, semantics/pragmatics and orthography. Tone is a complication and so is having a lot of homonyms. I once read that the syllable /i/ has 87 different meanings in Chinese. Is that more or less than a complication of having 17 different forms of the number 2? French has the phenomena of elision and liason and in Spanish the pronunciation of some words depends on the preceding word - those are all complications which a learner has to master.

Complexity is any event difficult to measure. Just keeping to verbs, we may be able to say that Ancient Greek is more complex than Latin because it has more forms, but how do we compare Latin with Spanish when the latter has fewer forms but more irregularities than the former? Is a language with 20 cases and no prepositions more complex than a language with no cases and 20 prepositions? Is the requirement to get words in the right order not just as much a complexity as getting the right endings on words?

As I said in anotther thread, an Ancient Roman learning English may feel it is complex because it involves:

· An obsession with specifying the definiteness of nouns.

· The not entirely straightforward way questions and negative statements are formed.

· The continuous and affirmative forms of verbs.

· The abundance of phrasal verbs whose meaning cannot always be guessed from their components.

· The need to get words in the right order to convey meaning.

· The inability to convey degrees of emphasis by word order.

· A general feeling that the language lacks precision and leaves too much to context.

Report

 

[jasio]

That assumes that complexity involves only morphology and ignores syntax, not to mention phonology, semantics/pragmatics and orthography. Tone is a complication and so is having a lot of homonyms. I once read that the syllable /i/ has 87 different meanings in Chinese. Is that more or less than a complication of having 17 different forms of the number 2?

Although a few years ago there were discussion on the Net claiming that Polish was the most difficult language globally, I do not go that far. I only claim that the modern Slavic languages are more complicated than most modern Western languages. I won't even claim "all modern Western languages", because I know nothing about Basque or Celtic languages.

how do we compare Latin with Spanish when the latter has fewer forms but more irregularities than the former?

Depending on how many is "more". And what do you understand as "irregularities": "o" -> "ue" for L-class irregular verbs is in fact a minor thing. So are spelling changes resulting from a bit awkward Spanish orthography - but it's a complication of the orthography per se, not an 'irregularity'.

Anyway, CIA evaluated it, at least for the English speakers, so yes - it's a mixture of a difficulty and deviation from English. But at least it's already done. In short: most Romance and Germanic languages are in the category 1. German is category 2 (it depends on the source actually), Slavic languages, Icelanding (as probably the only Germanic language) and most other languages are category 3, and Arabic, Chinese, Japanese and Korean are category 4.
https://cms.qz.com/wp-content/uploads/2017/11/how-far-will-language-study-take-you-_mapbuilder.png

Is a language with 20 cases and no prepositions more complex than a language with no cases and 20 prepositions?

Depending on how regular is the use of the prepositions and cases - both in terms of morphology and semantics.

Is the requirement to get words in the right order not just as much a complexity as getting the right endings on words?

I learned this requirement as a second language and believe me, it's peanuts. Sometimes I make mistakes in more complex situations, because I'm used to a more free syntax, but it's not something I would call complication - because it's regular. Asking questions by inversion is regular. Negation using an auxiliary verb is regular. In general, compound tenses are simpler than simple tenses - because they are regular. Learn several irregular forms of "to be" and "to have" - and you have 95% complications done, because verb forms used in compound tenses are mostly regular. Simple future tense in English is 100% regular - because it consists of a single auxiliary vern and the basic form of the verb. Present continuous is 99% regular - the only iiregularity is the auxiliary verb. OK, present perfect may cause some issues because most irregular verbs are commonly used.

 

[Hulalessar]

Difficulty is relative. The CIA evaluation is made from the perspective of native English speakers. Romance languages may be easier to learn than Slavic, but the main reason is that a huge amount of vocabulary is the same. I learned Russian at school along with French, Spanish and Latin. I did not find the grammar of Russian any more complicated than the others. A significant difficulty though is that you get no help with the vocabulary. I am not sure I ever quite got to grips with the perfective/imperfective thing, but then I do not always use the right form of the verb when speaking Spanish. The unpredictable stress was something of a problem. However, I was able to go from zero to reading Pushkin and Chekhov in two years. (Today I can hardly hold a conversation in Russian, but that is because since leaving school I have never used it except to order breakfast on a train in Romania in 1973. If you learn a lot in a short period you lose it if you don't use it.)

There is also a psychological aspect. If you tell someone something is difficult they will probably find it difficult. Russian is perceived to be difficult because it has a different script. English as a foreign language is typically started young and studied over a long period. Complications are taken one at a time and you get a lot of practice. You may start off saying "I do not must" but you will eventually get it right.

You also cannot ignore the fact that a huge number of people speak fluent Russian as a second language. Necessity is often the mother of second language acquistion. People will acquire a language if they really need to and they do not worry about difficulty or complexity. Difficulty and complexity do not come into it when children acquire their native language. Russian children use the right cases just as Spanish use the subjunctive when it is needed. There is nothing obvious about when a language which has articles uses them as the rules vary from language to language. If you look at the rules for any language they are quite involved, but small children follow them. Perceptions of difficulty and complexity come down to how different a language is from your own.

Although perceptions of complexity may be biased, complexity itself is not relative. The problem is though in devising a way to measure it. What exactly are you going to measure, how are you going to measure it and what weight are you going to give to each thing you measure? One thing you can be sure of is that linguists are never going to agree.

 

[Awwal12]

As I said in anotther thread, an Ancient Roman learning English may feel it is complex because it involves:

· An obsession with specifying the definiteness of nouns.

It wouldn't be complex by itself weren't the very cathegory of definiteness arbitrary enough. I often get the feeling that the native speakers just have made some secret agreement behind our backs about definiteness of every possible kind of noun groups. And as if it wasn't enough, encoding of definiteness in English is very complex by itself (the tripartite system of articles with a long, long list of specific cases and possible border situations again).