• I walked wide-open field-ward where John slaughtered the goat

• The wide-open field where John slaughtered the goat is pretty.

• I walked to the wide open field where John slaughtered the goat

• I walked to the wide open sheep grazing field-ward where John slaughtered the goat.

• I walked from the wide open sheep grazing field where John slaughtered the goat to the store.

• I walked the wide open sheep grazing field-from where John slaughtered the goat store-ward.

The case-introducing/case-removing noun-phrase scan is really annoying. It is computationally taxing, and I believe that it creates a linguistic pressure to remove cases from a language, and replace these with prepositions.

  When there are no cases, you have effortless noun-phrase embedding--- you just change the preposition, which always appears at the beginning.

I personally witnessed a case-shedding event, in modern Hebrew. In the Bible, you would always say "Halachti le-beyto", "I went to-his-house" to mean "I went to his house", using a possessive marker on the word "bayit" (house). But modern Hebrew speaker will always say "Halachti la-bayit shelo", "I went to the house of him", precisely because the word "shel" can be used to embed, as in "Halachti la-bayit shel ha-ach hagadol sheli" "I went to the house of my big brother".

Biblical Hebrew has a "to" case, so that it says "I walked Schem-ward" ("Halachti Schem-a"). Modern Hebrew, despite prescriptivist admonitions, dropped this case entirely, so that all modern speakers use the recursion friendly: "I walked to shchem" ("Halachti le-Schem"). From my own native speaker intuition, I know that this is a consequence of the ubiquitous embedding in modern Hebrew. You don't say "I walked city-ward by the sea" in modern Hebrew, it is ungrammatical. you say "I walked to the city by the sea", with the exact same form as in English.

The Bible doesn't embed very much, and if it wanted to do this, it would say it in a pre-literate recursion unfriendly form "Halachti Schema, zu ha-'ir le-yad ha-yam"/"I walked city-ward, this is the city by the sea."

So the hypothesis is that recursive languages shed their cases as soon as most speakers begin to produce and transform multiply embedded sentences on a regular basis.

• I walked wide-open field-ward where John slaughtered the goat

• The wide-open field where John slaughtered the goat is pretty.

• I walked to the wide open field where John slaughtered the goat

compare the cased:

• I walked store-from-ly the wide open sheep grazing field-ward where John slaughtered the goat.

Store from-ly means "from the store" in a case approximation in English.

• I walked the wide open sheep grazing field-from-ly where John slaughtered the goat store-ward.

to the uncased:

• I walked from the store to the wide open sheep grazing field where John slaughtered the goat.

• I walked to the store from the wide open sheep grazing field where John slaughtered the goat.

There is no internal scan of the noun-phrases required in order to change their function. the words stay the same as the noun-phrase takes on different roles, only the preposition changes.

The case-introducing/case-removing noun-phrase scan is really annoying. It is computationally taxing, and I believe that it creates a linguistic pressure to remove cases from a language, and replace these with prepositions. When there are no cases, you have effortless noun-phrase embedding--- you just change the preposition, which always appears at the beginning.

I personally witnessed case-shedding events, in modern Hebrew. In the Bible, you would always say "Halachti le-beyto", "I went to-his-house" to mean "I went to his house", using a possessive marker on the word "bayit" (house). But modern Hebrew speaker will always say "Halachti la-bayit shelo", "I went to the house of him", precisely because the word "shel" can be used to embed, as in "Halachti la-bayit shel ha-ach hagadol sheli" "I went to the house of my big brother".

In addition to a possessive marker, which was shed, Biblical Hebrew also has a "to" case, so that it says "I walked Schem-ward" ("Halachti Schem-a"). Modern Hebrew, despite prescriptivist admonitions, also dropped this case entirely, so that all modern speakers use the recursion friendly: "I walked to shchem" ("Halachti le-Schem"). 

From my own native speaker intuition, I know that this is a consequence of the ubiquitous embedding in modern Hebrew. You don't say "I walked city-ward by the sea" in modern Hebrew, it is ungrammatical. you say "I walked to the city by the sea", with the exact same form as in English.

The Bible doesn't embed very much, and if it wanted to do this, it would say it in a pre-literate way that suggests recursion is completely alien to the author: "Halachti Schema, zu ha-'ir le-yad ha-yam"/"I walked city-ward, this is the city by the sea."

So the hypothesis is that recursive languages shed their cases as soon as most speakers begin to produce and transform multiply embedded sentences on a regular basis. This happens at different times for different languages.

Given that Piraha (and Warlpiri) do not allow embedded clauses, and Piraha does not allow recursion at all, clause embedding is not universal to all languages. It is reasonable to suppose that it does not appear in full blown form in pre-literate societies. So unlimited embedding of clauses is something that post-dates writing and evolved along with it. This whole question presupposes this thesis.

To support this, I will give the evidence of the Hebrew bible (since the only ancient language I can read is Hebrew). From reading a few books of the old testament, and translating them to English, I can attest that the grammatical complexity of the old testament does not give any clear evidence of a grammar which allows Charles Dickens or Henry James recursive styles. The oldest parts never go more than one embedded clause deep, and there are grammar errors which suggest that the authors had a problem with complex sentences.

(EDIT: I was told by a classicist that this is not so pronounced in ancient greek and latin. A classics student I asked told me that Homer's work is recursive, although not as complex as later writing. He claimed that the later latin writers, he named Cicero and Thucidedes, are about equally recursive as Shakespeare and Dickens. This was interesting, because their recursive sentences occur in heavily cased languages, in ancient times.)

I am not asking about this, I just ask you to accept this hypothesis for the sake of making sense of this question. Assuming that grammatical recursion evolved in the historical period, one can ask: what effects did the introduction of recursive grammar have?

I believe that the major effect of the introduction of recursion is to disfavor the complex case systems typical of ancient and pre-literate languages, in favor of using separate words like "to" and "of". While it is easy enough to add a case marker to a stand-alone word, like house, so as to say "I walked house-ward" (in an approximation to a case in English), this construction does not work when you replace "house" by "place where my mother in law was born". You can say "I walked to the place where my mother in law was born", but you can't say "I walked (place where my mother in law was born)-ward".

Cases and clause recursion don't play well, for the same reason that free word order and recursion don't play well. If you can rearrange the words in a sentence, you can't make clauses nest, because you just can't rearrange words outside of clause boundaries without wrecking the embedded clause boundaries, and you need clear start and end markers for clauses, which are not provided without the appropriate stand-alone function words "to","in", "of", "which","that", and so on.

This hypothesis makes the following predictions, and these are the questions I have:

• Cases should be most diverse and pronounced in non-embedding languages, like Piraha. This is true of Piraha. Is it true of other non-embedding languages?
• Languages that embed early in history should shed their cases gradually, as embedding takes over. This means that the most cased languages today should have a recent writing system, and the least cased languages should have an ancient writing system. For Russian and English, this works. Are there counterexamples?
• Cases should be gradually more disfavored with time, as speakers internalize embedding, replacing the case markers with the function words. Modern Hebrew is a stunning example: in the Bible, you would alwayssay "Halachti le-beyto" to mean "I went to his house", using a possessive marker on the word "bayit" (house). But modern Hebrew speaker will always say "Halachti la-bayit shelo", "I went to the house of him", precisely because the word "shel" can be used to embed, as in "Halachti la-bayit shel ha-ach hagadol sheli" "I went to the house of my big brother". The Bible doesn't embed very much. Are there opposite events where the case system was strengthened? Perhaps after a loss of literacy?

The mainstream view of linguists is that case-shedding is a kind-of degenration of language with time, and that it occurs as speakers lose the quality of excellence of the early language. You hear this piffle from classicists when they discuss the Latin case system. I am saying that the cases disappear because the language is developing recursion, not because the speakers are getting dumber. They are getting smarter.

• Is there any recursion in ancient latin? How many levels deep? How does it work?
• Is there any evidence for this degenration idea?

(EDIT: It seems that the answer for ancient Latin is that the recursion is as developed as in modern languages.)

I am sorry for not separating the questions, but they are all about the same thing, and feel free to answer only one, any insight is welcome. I am not sure if it is in the linguistics literature, or if it is original to me. I think this idea is original, but please disabuse me of this belief if I am wrong.

EDIT: In response to comments

I got a downvote and weird comments, so I would like to make the hypothesis clearer. Consider the Hebrew sentence "and he walked to Shechem". In the Bible, it is "Ve-Halach Shchem-a" (and-(walked-he) Shechem-ward). In modern Hebrew, nobody would use this expression. They say "Ve-Halach le-Shchem" (and-(walked-he) to-Shechem).

What happened? The post-modifier "a" attaching to Shechem is replaced by a pre-modifier. Why does it happen? I know why--- because if you wanted to say "And he walked to the large mountain by the sea", you would say "Ve-halach la-har hagadol leyad ha-yam" (And-walked-he to-the-mountain the-big by the-sea). You could not say "Ve halach har-a hagadol leyad ha-yam", using the post modifier, without cluttering up the clause "har hagadol leyad ha-yam" with an extra syllable in the middle. By the way, if the Bible ever said such embedded things, it would say it this way:

"Vehalach har-a, ha-har hagadol leyad hayam". (And walked-he to-the-mountain, the-mountain the-big beside the sea.--- And he walked to the mountain, the big mountain beside the sea.)

It would use the post-modifier to make the mountain into the proper case, but it would then clarify what the mountain meant in a separate phrase that isn't really embedded. The case system is mucking up the embedding.

Nobody uses "mountain-ward" in modern Hebrew, except for some idiomatic phrases, like "halachti ha-baita" (I walked home-ward). The case modifier is replaced with what is essentially a stand-alone function word "le" or "la" (meaning "to" or "to the").

Why do you prefer the pre-modifier rather than the post-modifier? I strongly believe that the reason is that if you are doing Chomskian transformations which move phrases to different roles, you want to be able to move the whole phrase somewhere else as a unit, without mucking around inside to remove stupid dangling case-modifiers. So you prefer the function word "le" or "la", which functions exactly the same as the English "to". It comes at the beginning, and you just lop it off to make a stand-alone phrase that can be moved to another position in a different sentence.

To make the example clear, I will pretend that English can do the same thing. Suppose that English was able to say "I walked mountain-ward", using a modifier "ward" on words that tells you that you are walking to the mountain. Then you could say

"I walked mountain-ward with the statues of presidents carved into it"

or you could say

"I walked to a mountain with the statues of presidents carved into it"

The first form has a stupid case modifier in the middle of the clause that needs to be excised if you want to do a transformation. For example, if you want to say

"The mountain with the statues of presidents carved into it is pretty"

You would need to change mountain-ward into mountain, which requires mucking around inside the clause. This is much more intuitive when the function stuff occurs before the word, as a pre-modifier, rather than after the word, as a post-modifier.

So languages that deal with embedding on a regular basis like to have the syllable signifying "to" before the word, not after. In this position, it can drop off and become a separate word without any difficulty, and after this happens, you would say that the case has been lost, and a function word is gained.

In hebrew "le" is not a separate word in writing, but it might as well be, considering how it is used. It has the same exact function as the "to" in English, but it took over the "a" (-ward, as in homeward) post-modifier so much, that it is jarring to read the Bible and read "Lech Shchema ve-tagur sham." (Go Shchem-ward and live there). It seems obvious to me, considering how fast this happened (in the last 100 years) that the reason is that recursive clause embedding is common in modern Hebrew, and essentially nonexistent in Biblical hebrew.

The possessive 's of English is perhaps another reasonable example, but it is strange, in that you can apply it to clauses. I don't care about this weird exception. It seems that the rule is that when you have embedding, you prefer modifiers to happen at the beginning, and then they might as well drop off and become stand-alone function words.

This phenomenon of replacing post-modifiers or word-modifiers for case with a universal function word should happen to languages as they acquire recursive clause embedding. I want to know if this is true elsewhere than Hebrew.

EDIT: Explaining this in English

I will make up a case system for English, by replacing function words, so as to not have to write Latin examples.

verb modifiers will be as follows:

I walked to the store : I walked the-store-ward
I walked from the store : I walked the-store-from-ly
I walked on the beach : I walked the-beach-on-ly
I walked with my friend: I walked friend-of-mine-with-ly

etc., using the function word plus "ly" as the case word for a verb-attaching adjective-like (or argument like) phrase, except for "to-ly" which is "ward".

My bag with a zipper: Bag-of-mine zipper-with-ish
My coat on the desk : Coat-of-mine the-desk-on-ish

etc., using "ish" and the function word as the case marking for a adjective phrase

Then taking a sentence with embedding:

• I walked to the store by the sea
• I walked the-store-to-ly the-sea-by-ish

The second expression is easy to permute:

• I walked the-sea-by-ish the-store-to-ly

But when doing a transformation, like asking

• Is the-store by the sea green?
• Is the-store the-sea-by-ish green?

I have to take the phrase "the store-to-ly the-sea-by-ish" and transform it by scanning to remove the "to-ly" from the store. In the phrase "the store-to-ly the-sea-by-ish", "the store" is the leading noun, and it gets the case marker.

the This doesn't look too hard in this example, but if the leading noun is buried deep in the noun phrase, it becomes more difficult:

• I walked (sea-by-ish, which is next Greece-to-ish, store-to-ly which the Italy-of-ish health-department closed in Lord-of-ours-of-ish the-year 1999).
• I walked (to the store by the sea which is next to Greece, which the Italian health department closed in the year of our Lord 1999).

I enclosed the relevant phrase in parentheses in both examples (the intended parsing is that the sea is next to Greece, not the store), and italicized the cased leading noun, which in the example is "store-to-ly".

Now if I want to transform the phrase "the store by the sea which is next to Greece, which the Italian Health department closed in the year of our Lord 1999." into a question about whether this store is still closed, I have to take the phrase in parentheses, scan inside this phrase past "sea-by-ish, which is next Greece-to-ish," to find "store-to-ly" and then transform that token into "store". The transformed question is

• Is the (sea-by-ish, which is next Greece-to-ish, store which the Italy-of-ish health department closed in Lord-of-ours-of-ish the-year 1999) still closed?

You need to scan the phrase to remove the case markers from the leading noun. In standard English, no scan is ever required for a transformation--- you just move the entire phrase:

• Is (the store by the sea which is next to Greece, which the Italian health department closed in the year of our Lord 1999) still closed?

Notice that the parenthesized NP is exactly the same as before, just dropping the initial "to the". This makes transformations easy, even with deep embedding, and it is also the definition of what makes a language uncased. So uncased is recursion friendly, and cased is not.

I believe that the desire to make phrases movable without modification during transformations is the driving force for the phenomenon of case-shedding in languages.

The examples above, applying the case-markings to the main noun in the noun phrase, is how it works in Latin.

I edited this question in response to Karlsson's paper, "Constraints on Multiple Center-Embedding of Clauses" (Journal of Linguistics 43 (2), 2007, 365-392), linked here: http://www.ling.helsinki.fi/~fkarlsso/ceb5.pdf .

Context

Given that Piraha (and Warlpiri) do not allow embedded clauses, and Piraha does not allow recursion at all, clause embedding is not universal to all languages. As Karlsson persuasively argues, full blown recursion developed in the historical period. I will assume this here, and I will not repeat Karlsson's arguments.

In the Hebrew bible, there is no more than one embedded clause in any given sentence, consistent with Karlsson's bound for pre-literate societies, and there are grammar errors which suggest that the authors had a problem with complex sentences. As Karlsson explains, Greek and Latin writers developed full blown center-recursion explicitly somewhere around the 1st and 2nd centuries BC, and there is no evidence that recursion existed before then in any form

Hypothesis

As recursive embedding becomes common, it is possible that speakers prefer constructions which allow embedded clauses to take on different roles with no internal modifications. If you say

• I walked home-ward quickly

and you try to modify home to the embedded noun-phrase "the wide-open field where John slaughtered the goat", you say

• I walked wide-open field-ward where John slaughtered the goat

this requires inserting a syllable in the middle of the noun-phrase "wide-open field where John slaughtered the goat". If you want to move the noun-phrase to the subject position of a different sentence, you need to scan the interior of the noun-phrase to remove the case-modifier:

• The wide-open field where John slaughtered the goat is pretty.

If you have a stand-alone preposition, no scan is required

• I walked to the wide open field where John slaughtered the goat

the words past "to" are exactly the same as when the noun-phrase is in the subject role. If the preposition can appear at the beginning of the phrase, it gives a clear "push" indicator for a mechanical parser, and it allows for trivial transformation of phrases to different roles.

• I walked to the wide open sheep grazing field-ward where John slaughtered the goat.

• I walked from the wide open sheep grazing field where John slaughtered the goat to the store.

• I walked the wide open sheep grazing field-from where John slaughtered the goat store-ward.

The case-introducing/case-removing noun-phrase scan is really annoying. It is computationally taxing, and I believe that it creates a linguistic pressure to remove cases from a language, and replace these with prepositions.

When there are no cases, you have effortless noun-phrase embedding--- you just change the preposition, which always appears at the beginning.

I personally witnessed a case-shedding event, in modern Hebrew. In the Bible, you would always say "Halachti le-beyto", "I went to-his-house" to mean "I went to his house", using a possessive marker on the word "bayit" (house). But modern Hebrew speaker will always say "Halachti la-bayit shelo", "I went to the house of him", precisely because the word "shel" can be used to embed, as in "Halachti la-bayit shel ha-ach hagadol sheli" "I went to the house of my big brother".

Biblical Hebrew has a "to" case, so that it says "I walked Schem-ward" ("Halachti Schem-a"). Modern Hebrew, despite prescriptivist admonitions, dropped this case entirely, so that all modern speakers use the recursion friendly: "I walked to shchem" ("Halachti le-Schem"). From my own native speaker intuition, I know that this is a consequence of the ubiquitous embedding in modern Hebrew. You don't say "I walked city-ward by the sea" in modern Hebrew, it is ungrammatical. you say "I walked to the city by the sea", with the exact same form as in English.

The Bible doesn't embed very much, and if it wanted to do this, it would say it in a pre-literate recursion unfriendly form "Halachti Schema, zu ha-'ir le-yad ha-yam"/"I walked city-ward, this is the city by the sea."

So the hypothesis is that recursive languages shed their cases as soon as most speakers begin to produce and transform multiply embedded sentences on a regular basis.

Predictions/Questions

This hypothesis makes the following predictions, and these are the questions I have:

• Cases should be most diverse and pronounced in non-embedding languages, like Piraha. This is true of Piraha. Is it true of other non-embedding languages?
• The most cased languages today should have a recent writing system, and the least cased languages should have an ancient writing system. For Russian and English, this works. Are there counterexamples?
• Are there opposite events where the case system was strengthened? Do these correspond to a loss of literacy?

It would use the post-modifier to make the mountain into the proper case, but it would then clarify what the mountain meant in a separate clause that isn't really embedded. The case system is mucking up the embedding.

Why do you prefer the pre-modifier rather than the post-modifier? I strongly believe that the reason is that if you are doing Chomskian transformations which move clauses to different roles, you want to be able to move the whole clause somewhere else as a unit, without mucking around inside to remove stupid dangling case-modifiers. So you prefer the function word "le" or "la", which functions exactly the same as the English "to". It comes at the beginning, and you just lop it off to make a stand-alone clause that can be moved to another position in a different sentence.

(EDIT: I was told by a classicist that this is not so pronounced in ancient greek and latin. A classics student I asked told me that Homer's work is recursive, although not as complex as later writing. He claimed that the later latin writers, he named Cicero and Thucidedes, are about equally recursive as Shakespeare and Dickens. This was interesting, because their recursive sentences occur in heavily cased languages, in ancient times.)

(EDIT: It seems that the answer for ancient Latin is that the recursion is as developed as in modern languages.)

It would use the post-modifier to make the mountain into the proper case, but it would then clarify what the mountain meant in a separate phrase that isn't really embedded. The case system is mucking up the embedding.

Why do you prefer the pre-modifier rather than the post-modifier? I strongly believe that the reason is that if you are doing Chomskian transformations which move phrases to different roles, you want to be able to move the whole phrase somewhere else as a unit, without mucking around inside to remove stupid dangling case-modifiers. So you prefer the function word "le" or "la", which functions exactly the same as the English "to". It comes at the beginning, and you just lop it off to make a stand-alone phrase that can be moved to another position in a different sentence.

EDIT: Explaining this in English

I will make up a case system for English, by replacing function words, so as to not have to write Latin examples.

verb modifiers will be as follows:

I walked to the store : I walked the-store-ward
I walked from the store : I walked the-store-from-ly
I walked on the beach : I walked the-beach-on-ly
I walked with my friend: I walked friend-of-mine-with-ly

etc., using the function word plus "ly" as the case word for a verb-attaching adjective-like (or argument like) phrase, except for "to-ly" which is "ward".

My bag with a zipper: Bag-of-mine zipper-with-ish
My coat on the desk : Coat-of-mine the-desk-on-ish

etc., using "ish" and the function word as the case marking for a adjective phrase

Then taking a sentence with embedding:

• I walked to the store by the sea
• I walked the-store-to-ly the-sea-by-ish

The second expression is easy to permute:

• I walked the-sea-by-ish the-store-to-ly

But when doing a transformation, like asking

• Is the-store by the sea green?
• Is the-store the-sea-by-ish green?

I have to take the phrase "the store-to-ly the-sea-by-ish" and transform it by scanning to remove the "to-ly" from the store. In the phrase "the store-to-ly the-sea-by-ish", "the store" is the leading noun, and it gets the case marker.

the This doesn't look too hard in this example, but if the leading noun is buried deep in the noun phrase, it becomes more difficult:

• I walked (sea-by-ish, which is next Greece-to-ish, store-to-ly which the Italy-of-ish health-department closed in Lord-of-ours-of-ish the-year 1999).
• I walked (to the store by the sea which is next to Greece, which the Italian health department closed in the year of our Lord 1999).

I enclosed the relevant phrase in parentheses in both examples (the intended parsing is that the sea is next to Greece, not the store), and italicized the cased leading noun, which in the example is "store-to-ly".

Now if I want to transform the phrase "the store by the sea which is next to Greece, which the Italian Health department closed in the year of our Lord 1999." into a question about whether this store is still closed, I have to take the phrase in parentheses, scan inside this phrase past "sea-by-ish, which is next Greece-to-ish," to find "store-to-ly" and then transform that token into "store". The transformed question is

• Is the (sea-by-ish, which is next Greece-to-ish, store which the Italy-of-ish health department closed in Lord-of-ours-of-ish the-year 1999) still closed?

You need to scan the phrase to remove the case markers from the leading noun. In standard English, no scan is ever required for a transformation--- you just move the entire phrase:

• Is (the store by the sea which is next to Greece, which the Italian health department closed in the year of our Lord 1999) still closed?

Notice that the parenthesized NP is exactly the same as before, just dropping the initial "to the". This makes transformations easy, even with deep embedding, and it is also the definition of what makes a language uncased. So uncased is recursion friendly, and cased is not.

I believe that the desire to make phrases movable without modification during transformations is the driving force for the phenomenon of case-shedding in languages.

The examples above, applying the case-markings to the main noun in the noun phrase, is how it works in Latin.