Folks,
Lexical semantics is a large field and there are various positions specialists
take on exactly how word-meanings are best to be characterized. For example,
most scholars (not all), argue that there is no simple verb 'to give' but that
this English verb is characterized by a representation along the lines of:
Anna gave Max a book.
give: lambda z lambda y lambda x lamba e act(x) & become poss(y,z)(e)
('lambda' is of course the lambda operator)
In other words, any verb, in this case 'give' is broken down into more basic
components. No language is required to have a verb that is exactly like the
English verb 'to give' but if it does, it must be composed of these finer
predicates, so that the triadic semantics of 'to give' (English) is derivative,
not basic (though the combinations of these basic predicates in this form will
in fact produce a di-transitive or "triadic" syntax).
Some linguists would refer to the number of lexical arguments as the valency of
the verb and the number of syntactic arguments as the transitivity of the verb
(noun, etc).
And this can vary radically across languages.
For example, in the Piraha language that I have worked on for decades, there
are only about 90 or so distinct verb roots (which are not to be confused with
verb stems, in turn not to be confused with verbs, and not to be confused with
lexical representations). So to produce a verb like 'bring back' (corresponding
roughly to a single verb such as 'return', as in 'return the screwdriver when
you're finished') in Piraha the actual verb might be: 'go-turn-carry-aspectual
distinction affixes...' (i.e. a verb stem composed of several verb roots plus a
number of affixes playing derivative semantic roles).
In a language like English with an extremely simple verbal morphology (maximum
of five forms - sing, sang, sung, sings, singing) this is deceptively easy. In
Spanish a verb would have 30-50 forms. But in Piraha (not uncommon for
polysynthetic languages of the Americas) each verb can have as many as 65,000
forms (sixty-five thousand). And there simply is no way to compare predicates
ilke "give" one-to-one with any Piraha verb.
If we consider a basic English-conceived/interpreted predicate like 'give' then
of course it is difficult to imagine that it wouldn't have three arguments,
e.g. a giver, a thing given, and a recipient. But this simple conception does
not manifest itself in all languages. So for example, abstracting away from the
much greater complexity of Piraha verbal morphology, there simply is no simple
verb 'to give' in the language. There is a combination of verb roots that
means, roughly, 'x transfers y to z with the expectation that z will transfer b
to x' That is, the concept of giving requires reciprocity in Piraha (a
hunter-gatherer culture).
In the case of Amele that I mentioned earlier, there is NO verb 'to give.' One
can find, for example, 'John apple Bill' with no verb and in the right context
imply that 'John gave an apple to Bill.' But there is no actual verb translated
'give' in the language.
All languages are severely *underdetermined* meaning that the predicates of the
language do not fully specify meanings or even literally specify meanings. Full
meanings are supplied by the cultural act of interpretation.
These are reasons that I favor the triadic system of semiotics of CSP to the
dyadic system of Saussure. But the argumentation is subtle and detailed. And
the semiotics has to fit the linguistics - it does not substitute for the
linguistics.
As far as syntax goes, syntax never, ever fully specifies all the imaginable
connections between predicates and arguments. Syntax acts as a filter on
potential meanings, guiding the interpretation process to avoid certain
interpretations. This active, dynamic role of interpretation is why the
Saussurian system cannot work and why I favor Peirce's work.
Now to the syntax. If we were to assume that there are three symbols (without
worrying about lexical decomposition), e.g. "John," "give," "apple", "Mary," we
can think of various ways of combining them. The simplest way would be simple
linear concatenation: "John give Mary apple." Another would be hierarchy +
concatenation, e.g. "[John [[give Mary] apple]]" and the final would be
hierarchy + recursion.
Peirce's semiotics is semantically recursive, though the syntax is of almost no
concern to that work. Chomsky's grammatical proposals require both syntactic
and semantic recursion. My own work requires Peircean recursion, while allowing
syntactic recursion to be optional across languages.
This would all take a lot more space for me to unpack for non-specialists (i.e.
non-linguists). Peirce's discussions of predicates is little more than a first
pass. This doesn't mean he was wrong, but that verbal semantics is far more
complex than he understood, because it is based on a science that did not
exist, except perhaps in an extremely inchoate form that he helped to birth, at
the time he was writing.
Guessing at whether predicates are triadic or not by non-linguists is akin to
me, a non-engineer, telling my grandson, a bio-medical engineer, how I think
the best way to design an artificial heart would be.
I have referred many times to work of mine and others on these and related
subjects. I suggest reading a few of those (or just an introductory linguistics
text). Here are a couple of more references:
http://www.zas.gwz-berlin.de/fileadmin/mitarbeiter/wunderlich/Wunderlich_14_Lexical_Decomposition.pdf
<http://www.zas.gwz-berlin.de/fileadmin/mitarbeiter/wunderlich/Wunderlich_14_Lexical_Decomposition.pdf>
https://dash.harvard.edu/bitstream/handle/1/3353824/Huang_Lexical.pdf
<https://dash.harvard.edu/bitstream/handle/1/3353824/Huang_Lexical.pdf>
http://www.oxfordhandbooks.com/view/10.1093/oxfordhb/9780199541072.001.0001/oxfordhb-9780199541072-e-15
<http://www.oxfordhandbooks.com/view/10.1093/oxfordhb/9780199541072.001.0001/oxfordhb-9780199541072-e-15>
https://www.amazon.com/How-Language-Began-Humanitys-Invention/dp/0871407957
<https://www.amazon.com/How-Language-Began-Humanitys-Invention/dp/0871407957>
And then for the connoisseur, there are grammars of individual languages.
So there is no utility in statements like "I certainly do assume that ... the
relation of giving is irreducbly triadic." First, even in English it IS
reduceable. Second, there is considerable cross-linguistic variation in the
mapping between lexical semantics (i.e. with lexical decomposition) and syntax,
or, as I stated earlier, between valency and transitivity.
Thus the English verb "to give" is ditransitive, but not necessarily divalent.
Dan
> On Apr 20, 2019, at 5:37 PM, Helmut Raulien <[email protected]> wrote:
>
> list,
> I approve: languages and cultures are tools! They can be used as far as they
> are good for, but they don´t limit! Other tools are easily available for
> everybody, for Amele people too. Dyadic giving, m. a! They gave something
> away to the world, or they gave everything they could to a special person?
> Before missionaries arrived, and it worked?
> Best, helmut
>
> 20. April 2019 um 21:17 Uhr
> "Gary Richmond" <[email protected]>
> wrote:
> Jon, Dan, List,
>
> Jon quoted Dan, then commented:
>
> DE: By the way, predicates in natural languages can vary for cultural
> reasons. For example, "to give" in most languages is triadic, but it is
> dyadic in Amele of New Guinea (as I discuss in Language: The Cultural Tool).
>
> JAS: That is interesting, but I assume we can still say that the relation of
> giving is irreducibly triadic, regardless of how it is expressed in any
> particular language.
>
> I certainly do assume that we can say "that the relation of giving is
> irreducibly triadic" and, this seems to me, to point again to the essential
> distinction between 'relation(ship)' and 'correlation' which Jon has been
> emphasizing in consideration of triadic relations. Perhaps this distinction
> also points to a difference in the subject matter of linguistics and
> semeiotic, whatever their overlaps may be.
>
> In any event, the exchange between Dan and Jon reminded me of the peculiar
> case of the language of the Trobriand Islanders, one I'd learned about
> decades ago in a college course which included analyses of certain cultural
> and even conceptual differences expressed through language. The Trobriand
> Islanders language is especially interesting for several reasons.
>
> The Trobriand peoples speak Kilivila
> <https://en.wikipedia.org/wiki/Kilivila%E2%80%93Louisiades_languages>, though
> various different dialects of it are spoken amongst each different tribe. It
> is an Austronesian language
> <https://en.wikipedia.org/wiki/Austronesian_language>, although [it] has the
> distinction of having a complex system for classifying nouns
> <https://en.wikipedia.org/wiki/Noun>.
>
> In this regard and, for example, it would appear that Kilivila doesn't
> express 'causation' or, perhaps, one might say 'process', 'growth', etc. As
> the article quoted from would have it, there appear to be in some languages
> "non-lineal codifications of reality." As this appears as a characteristic of
> Kilivila, one reads:
>
> Drawing upon earlier work earlier work by Bronisław Malinowski
> <https://en.wikipedia.org/wiki/Bronis%C5%82aw_Malinowski>, Dorothy D. Lee
> <https://en.wikipedia.org/wiki/Dorothy_D._Lee>'s scholarly writings refer to
> "non-lineal codifications of reality". In such a linguistic system, the
> concept of linear progress of time, geometric shapes, and even conventional
> methods of description, are lost altogether or altered. In her example of a
> specific indigenous yam, Lee explains that when the yam moves from a state of
> sprouting through ripeness to over-ripeness, the name for each object in a
> specific state changes entirely - the description of the object at different
> states of development relates to wholly different perceptions of the object.
> Ripeness is considered a "defining ingredient" and thus once it becomes
> over-ripe, a yam is perceived as a new object altogether. The same perception
> pertains to time and geometric shapes.
> Lee, Dorothy D. B. (1950). Lineal and nonlineal codifications of reality.
> Psychosomatic Medicine march-april 1950. p. 89.
> https://en.wikipedia.org/wiki/Trobriand_Islands#Language
> <https://en.wikipedia.org/wiki/Trobriand_Islands#Language>
>
> While, of course, in nature there is process, growth, change, development,
> ripening, decay, etc., apparently language does not necessarily need to
> reflect that fact. I should note that I have since read somewhat different
> analyses of Kilivila than Lee's, but I have not been able to quickly locate
> them.
>
> Best,
>
> Gary
>
>
> Gary Richmond
> Philosophy and Critical Thinking
> Communication Studies
> LaGuardia College of the City University of New York
>
>
>
>
> <http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail>
> Virus-free. www.avg.com
> <http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail>
>
> On Sat, Apr 20, 2019 at 1:05 PM Jon Alan Schmidt <[email protected]
> <mailto:[email protected]>> wrote:
> Dan, List:
>
> DE: By the way, predicates in natural languages can vary for cultural
> reasons. For example, "to give" in most languages is triadic, but it is
> dyadic in Amele of New Guinea (as I discuss in Language: The Cultural Tool).
>
> That is interesting, but I assume we can still say that the relation of
> giving is irreducibly triadic, regardless of how it is expressed in any
> particular language. Do you disagree?
>
> DE: On another note, neurolinguistics research by Evelina Fedorenko and her
> colleagues at MIT's Brain and Cognitive Sciences Department shows that
> language networks in the brain are built primarily around words/symbols, with
> syntax parasitic on words, which I independently predict in my book and
> recent work on the origin of language. Also Steve Piantadosi (Berkeley
> Psychology) has shown that all the memory needed for much of language is
> about 1.5 megabytes, a small amount, with the amount of memory required for
> syntax negligible, more evidence that symbols drive syntax.
>
> Could you please elaborate on what you mean by "syntax parasitic on words"?
> I am neither a linguist nor a cognitive scientist--and I did not stay at a
> Holiday Inn Express last night--but that interpretation of the evidence seems
> backwards. Consider this excerpt from the article about Piantadosi's
> research that you linked
> <https://www.sciencedaily.com/releases/2019/03/190327134547.htm> in a
> subsequent post.
>
> For example, when presented with the word "turkey," a young learner typically
> gathers bits of information by asking, "Is a turkey a bird? Yes, or no? Does
> a turkey fly? Yes, or no?" and so on, until grasping the full meaning of the
> word "turkey."
>
> The young learner in this example must already have mastered basic syntax in
> order to pose his/her questions in the first place, and then understand the
> responses--which help make the word "turkey" more determinate in his/her own
> mind, like adding Spots to a branching Line of Identity. This is presumably
> why symbols require far more memory than syntax--they have to be situated in
> relation to all other symbols, while the same syntax applies to a vast range
> of different circumstances, as the isomorphic Existential Graphs for the
> triadic relations of giving and determining reflect.
>
> Moreover, what do you make of the following comments by Peirce regarding the
> priority of syntax?
>
> CSP: Take the Proposition "Cain killed Abel." This is identically the same
> Proposition as "Abel was killed by Cain": It is only the grammatical dress
> that is different. Other things being equal, everybody will prefer the
> former. Why? Because it is simpler; but why is it simpler? Because in
> putting the cause before the effect, it in that respect diagrammatizes the
> truth. What are the Subjects of this Proposition. Cain, first: that is not
> only a Subject of the Proposition, but is the principal Subject of the
> Assertion which a historian would naturally make. But in the Proposition
> Cain and Abel are, as Subjects, on one footing precisely (or almost
> precisely, for Cain is preponderant in causality). But besides these,
> "killed" = committed murder upon, is a third Subject, since no study of the
> words alone, without extraneous experience, would enable the Ad[d]ressee to
> understand it. What, then, is left to serve as Predicate? Nothing but the
> flow of causation. It is true that we are more acquainted even with that in
> Experience. When we see a babe in its cradle bending its arms this way and
> that, while a smile of exultation plays upon its features, it is making
> acquaintance with the flow of causation. So acquaintance with the flow of
> causation so early as to make it familiar before speech is so far acquired
> that an assertion can be syntactically framed, and it is embodied in the
> syntax of every tongue. However, it is not because of this physiological
> fact, that it becomes proper to draw the line between Subjects and Predicate
> here; neither is it because of the psychical fact that human minds naturally
> think in a way broadly (i.e. a little) similar to the forms of syntax; nor is
> it even because of the metaphysical truth, that the order of syntax is the
> law of Time and of Becoming. This is proved by the facts, first, that it is
> necessary that the Reasonings by which we discover and defend the orders of
> Causation, of human thought, of time, and of becoming, themselves presuppose
> the recognition of the corresponding order in syntax; and secondly, by this,
> that it has not been the order [of] time, or Causation, or the structure of
> the human mind, nor human anatomy and physiology that have, any or all of
> them, determined that that ought to be the order of syntax that in fact ought
> to be so, but precisely the contrary, it is the fact that the order of Syntax
> ought to be as in fact it ought to be that has determined, first, Real Being
> and Time to take the same form, and then, that it should become natural to
> the mind and should be the pattern of physical action. (R 664:10-13; 1910 Nov
> 26-27)
>
> It seems to me--and Peirce, apparently--that another reason why syntax
> requires little memory is simply because it is relatively natural, generally
> reflecting "the flow of causation" that we discover long before we achieve
> competence in any language. By contrast, symbols are ultimately arbitrary,
> requiring the acquisition of myriad habits for even rudimentary employment in
> thought and communication.
>
> In any case, with Peirce I see the primary distinction here as being between
> subjects that denote Objects (as Semes) and continuous predicates that
> signify Interpretants (within Propositions). There is no limit to the
> potential number of different words or other symbols that we could devise and
> memorize to serve as the former, but only a handful of syntactical
> configurations are sufficient to express the latter, since they represent
> indecomposable logical relations. We can stipulate the proper linear
> sequence of words for this purpose, often supplemented by additional words;
> or we can adopt even more iconic Signs, such as Spots with certain
> arrangements of Pegs attached to other Spots by means of Lines, all scribed
> on a two-dimensional Sheet.
>
> Regards,
>
> Jon Alan Schmidt - Olathe, Kansas, USA
> Professional Engineer, Amateur Philosopher, Lutheran Layman
> www.LinkedIn.com/in/JonAlanSchmidt
> <http://www.linkedin.com/in/JonAlanSchmidt> - twitter.com/JonAlanSchmidt
> <http://twitter.com/JonAlanSchmidt>
> On Fri, Apr 19, 2019 at 1:20 PM Dan Everett <[email protected]
> <mailto:[email protected]>> wrote:
> The most common word order among the world's 7000+ languages is Sub Obj Verb.
> But I have worked on languages in the Amazon that are SOV, VSO, VOS, OSV,
> SVO, and OVS. I have written full grammars of two of them. A propositions can
> serve as predicates of higher propositions in some languages. Syntactic
> variation is intense.
>
> On another note, neurolinguistics research by Evelina Fedorenko and her
> colleagues at MIT's Brain and Cognitivie Sciences Department shows that
> language networks in the brain are built primarily around words/symbols, with
> syntax parasitic on words, which I independently predict in my book and
> recent work on the origin of language. Also Steve Piantadosi (Berkeley
> Psychology) has shown that all the memory needed for much of language is
> about 1.5 megabytes, a small amount, with the amount of memory required for
> syntax neglible, more evidence that symbols drive syntax.
>
> Dan
> On Apr 19, 2019, at 2:14 PM, Edwina Taborsky <[email protected]
> <mailto:[email protected]>> wrote:
> Yes, and the location of the predicate in the sentence is also interesting.
> For example, in classical ancient Chinese, the verb was at the end [as it is
> in Latin]. I thought this implied that whatever was going on in the full
> sentence, i.e, subject and object, involved both as co-agents in the action.
> What was interesting is how this ancient Chinese format eventually gave way
> to the modern mode where the verb is between the subject and object.
>
> Edwina
>
> On Fri 19/04/19 1:42 PM , Dan Everett [email protected]
> <mailto:[email protected]> sent:
>
> By the way, predicates in natural languages can vary for cultural reasons.
> For example, "to give" in most languages is triadic, but it is dyadic in
> Amele of New Guinea (as I discuss in Language: The Cultural Tool).
>
> Dan
>
>
> <http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail>
> Virus-free. www.avg.com
> <http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail>-----------------------------
> PEIRCE-L subscribers: Click on "Reply List" or "Reply All" to REPLY ON
> PEIRCE-L to this message. PEIRCE-L posts should go to [email protected]
> . To UNSUBSCRIBE, send a message not to PEIRCE-L but to [email protected]
> with the line "UNSubscribe PEIRCE-L" in the BODY of the message. More at
> http://www.cspeirce.com/peirce-l/peirce-l.htm
> <http://www.cspeirce.com/peirce-l/peirce-l.htm> .
>
> -----------------------------
> PEIRCE-L subscribers: Click on "Reply List" or "Reply All" to REPLY ON
> PEIRCE-L to this message. PEIRCE-L posts should go to [email protected]
> . To UNSUBSCRIBE, send a message not to PEIRCE-L but to [email protected]
> with the line "UNSubscribe PEIRCE-L" in the BODY of the message. More at
> http://www.cspeirce.com/peirce-l/peirce-l.htm .
>
>
>
>
-----------------------------
PEIRCE-L subscribers: Click on "Reply List" or "Reply All" to REPLY ON PEIRCE-L
to this message. PEIRCE-L posts should go to [email protected] . To
UNSUBSCRIBE, send a message not to PEIRCE-L but to [email protected] with the
line "UNSubscribe PEIRCE-L" in the BODY of the message. More at
http://www.cspeirce.com/peirce-l/peirce-l.htm .
