3 ms·

The point there is that it's not necessarily obvious that sign languages would have phonology or morhphology or syntax, but once you've realized that that's tru

The point there is that it's not necessarily obvious that sign languages would have phonology or morhphology or syntax, but once you've realized that that's true it absolutely does follow that they are as expressive (in the technical sense) as spoken languages. I don't want to get into an argument about linguistic relativity and the Minimalist program on Hacker News, but the Minimalists believe that the expressiveness of languages stem from their syntax: once a language has two syntactic operations (called "merge" and "move", that all natural languages have[0]), it becomes as expressive as every other natural language. Where you hide the complexity of this expressiveness depends a lot on the language: some put it in the lexicon, others in the morphology, etc. But at the end of the day, if sign languages have a syntax, they fall into the same class of expressiveness as spoken languages. As linguists started studying how something that looked like phonology, morphology and syntax might manifest in sign languages, it became clear that sign languages did have a syntax. You can see something that imo looks like the "move" operation going on with topicalization in ASL and something like "merge" going on with negations and numerical incorporation, although I don't actually know enough about ASL to say for sure. [0] Again, I really don't want to argue about this but one linguist named Daniel Everett doesn't think this is true (he believes that a language called Piraha does not have recursion or, by implication, a merge operation). But he's probably wrong, and Andrew Nevins and David Pesetsky at MIT have a very compelling argument about why that is.

thaumasiotes 7y ago
Sure. But as I already said, syntax is not a domain that "traditionally only applied to spoken languages". Its application to sign languages differs in no way from its application to spoken languages. And syntax is the relevant domain for asking whether sign languages are as expressive as spoken languages are. So your original claim looks like this hypothetical one: > Pete Ricketts is the governor of Nebraska, so it is now widely acknowledged that the sum of the first n consecutive odd numbers is always a perfect square. The premise is true. The conclusion is true. But this is a terrible argument to make. > it's not necessarily obvious that sign languages would have phonology or morphology or syntax, but once you've realized that that's true it absolutely does follow that they are as expressive (in the technical sense) as spoken languages. This is wrong on both points: - By definition, sign languages do not have phonology, do have morphology, and do have syntax; you can reach those conclusions without needing to investigate any language in particular. - Many other information channels, such as, say, Python, or bee dancing, do have morphology and syntax (and don't have phonology), but are obviously not as expressive as spoken languages are.
- eindiran 7y ago
  Sign languages absolutely do have a phonology. From the first sentence of the Wikipedia page on phonology: > Phonology is a branch of linguistics concerned with the systematic organization of sounds in spoken languages and signs in sign languages. I am using "syntax" in a very precise sense, as tied to linguistics generally and the Minimalist program specifically. Having a syntax = having a grammar with a merge operation and a move operation that governs relationships of the word or morpheme level units in your language. By that definition syntax is a domain "traditionally only applied to spoken languages", with the one caveat that there are a few papers out there examining whether non-linguistic grammars have merge or move operations. Because it wasn't always clear that sign languages have morphemes or words (it became widely known during the 1970s and 1980s), the syntacticians that were developing early ideas in mainstream syntax weren't doing so with sign languages in mind. So as the linguistic dogma "all languages are equally expressive" was developed, people were often not thinking explicitly of sign languages. That "all languages (including sign languages) are equally expressive" came to be appreciated was precisely because linguists started to examine how syntax and morphology might apply sign languages. If you can cite a single paper published before 1970 that discusses the syntax of a sign language from the perspective of generative linguistics, I will recant my whole statement and acknowledge that there is nothing meaningful tying the first half of the problematic sentence to the second half and that my understanding of the history of linguistics was incorrect. The earliest example I am aware of was from 1974.
  - thaumasiotes 7y ago
    If you're going to define phonology that way, it doesn't make any sense to say that there are people studying how subfields of linguistics that traditionally apply only to spoken languages can be applied to sign languages as well, since -- by definition -- they all apply equally to spoken languages and to sign languages. That reduces the claim to "some linguists study sign languages", which is even less interesting. > I am using "syntax" in a very precise sense, as tied to linguistics generally and the Minimalist program specifically. Having a syntax = having a grammar with a merge operation and a move operation that governs relationships of the word or morpheme level units in your language. Similarly, this is a bizarre viewpoint to take. Syntax is a much older concept than that. The subfield of syntax, as "traditionally" applied to spoken languages, predates the definition you're trying to use here, by hundreds of years. And that's just talking about the subject that was literally named "syntax". Take a look at gloss A.4 here: http://www.perseus.tufts.edu/hopper/text?doc=Perseus%3Atext%3A1999.04.0057%3Aentry%3Dsu%2Fntacis http://www.perseus.tufts.edu/hopper/text?doc=Perseus%3Atext%... . You are not using "syntax" in a sense that linguistics in general would recognize. You're taking a very specific feature and, without justification, naming it "syntax" in spite of the existing use of the word -- and then claiming that people who use the word normally just don't understand you. Of course they don't! And the subfield of syntax, as traditionally understood by everyone, conceptually applies to any messages governed by any rules at all. Are you really claiming that Daniel Everett would say the Pirahã language has no syntax? Or would he say that it does have syntax, and its syntax differs from those of most natural languages in that it lacks a particular feature?

thaumasiotes 7y ago

Sure. But as I already said, syntax is not a domain that "traditionally only applied to spoken languages". Its application to sign languages differs in no way from its application to spoken languages. And syntax is the relevant domain for asking whether sign languages are as expressive as spoken languages are. So your original claim looks like this hypothetical one: > Pete Ricketts is the governor of Nebraska, so it is now widely acknowledged that the sum of the first n consecutive odd numbers is always a perfect square. The premise is true. The conclusion is true. But this is a terrible argument to make. > it's not necessarily obvious that sign languages would have phonology or morphology or syntax, but once you've realized that that's true it absolutely does follow that they are as expressive (in the technical sense) as spoken languages. This is wrong on both points: - By definition, sign languages do not have phonology, do have morphology, and do have syntax; you can reach those conclusions without needing to investigate any language in particular. - Many other information channels, such as, say, Python, or bee dancing, do have morphology and syntax (and don't have phonology), but are obviously not as expressive as spoken languages are.

eindiran 7y ago
Sign languages absolutely do have a phonology. From the first sentence of the Wikipedia page on phonology: > Phonology is a branch of linguistics concerned with the systematic organization of sounds in spoken languages and signs in sign languages. I am using "syntax" in a very precise sense, as tied to linguistics generally and the Minimalist program specifically. Having a syntax = having a grammar with a merge operation and a move operation that governs relationships of the word or morpheme level units in your language. By that definition syntax is a domain "traditionally only applied to spoken languages", with the one caveat that there are a few papers out there examining whether non-linguistic grammars have merge or move operations. Because it wasn't always clear that sign languages have morphemes or words (it became widely known during the 1970s and 1980s), the syntacticians that were developing early ideas in mainstream syntax weren't doing so with sign languages in mind. So as the linguistic dogma "all languages are equally expressive" was developed, people were often not thinking explicitly of sign languages. That "all languages (including sign languages) are equally expressive" came to be appreciated was precisely because linguists started to examine how syntax and morphology might apply sign languages. If you can cite a single paper published before 1970 that discusses the syntax of a sign language from the perspective of generative linguistics, I will recant my whole statement and acknowledge that there is nothing meaningful tying the first half of the problematic sentence to the second half and that my understanding of the history of linguistics was incorrect. The earliest example I am aware of was from 1974.
- thaumasiotes 7y ago
  If you're going to define phonology that way, it doesn't make any sense to say that there are people studying how subfields of linguistics that traditionally apply only to spoken languages can be applied to sign languages as well, since -- by definition -- they all apply equally to spoken languages and to sign languages. That reduces the claim to "some linguists study sign languages", which is even less interesting. > I am using "syntax" in a very precise sense, as tied to linguistics generally and the Minimalist program specifically. Having a syntax = having a grammar with a merge operation and a move operation that governs relationships of the word or morpheme level units in your language. Similarly, this is a bizarre viewpoint to take. Syntax is a much older concept than that. The subfield of syntax, as "traditionally" applied to spoken languages, predates the definition you're trying to use here, by hundreds of years. And that's just talking about the subject that was literally named "syntax". Take a look at gloss A.4 here: http://www.perseus.tufts.edu/hopper/text?doc=Perseus%3Atext%3A1999.04.0057%3Aentry%3Dsu%2Fntacis http://www.perseus.tufts.edu/hopper/text?doc=Perseus%3Atext%... . You are not using "syntax" in a sense that linguistics in general would recognize. You're taking a very specific feature and, without justification, naming it "syntax" in spite of the existing use of the word -- and then claiming that people who use the word normally just don't understand you. Of course they don't! And the subfield of syntax, as traditionally understood by everyone, conceptually applies to any messages governed by any rules at all. Are you really claiming that Daniel Everett would say the Pirahã language has no syntax? Or would he say that it does have syntax, and its syntax differs from those of most natural languages in that it lacks a particular feature?