Jones assumes a referential theory of meaning drawn from ethnoscience and logical positivism. He cites Dwight Read and me without recognizing that we argue for a very different, pragmatic, conception (Leaf 1971; 2007; 2009; Read 2001; 2007). Pragmatic theory is demonstrably more parsimonious, precise, and powerful.
Referential theories are set up in terms of words and their referents. Pragmatic theories are set up in terms of ideas and their uses. Jones offers a list of universal parameters for classifying what kinship terms refer to and justifies it by invoking the supposed success of OT in linguistics. Read and I, by contrast, have demonstrated (not just suggested) that terminologies are systems of interrelated definitions, and that their internal order derives from computational rules contained in the core of terms that define the “direct” relations around the user, or self. Referring to relatives is just one of the many uses of such conceptual systems.
Jones begins by distinguishing two levels of language: “ordinary” and “VIP.” Ordinary is “open class or lexical,” which “includes most nouns and verbs.” He describes VIP as “closed-class” but otherwise only by example. The examples are grammatical markers. He puts kin terms in the VIP class. It is a crucial step in the argument, and it is unjustifiable. Languages have many closed classes, including classes of nouns and verbs. Kinship terminologies do form closed systems, but not classes in this sense.
Jones then argues that grammar uses spatial and temporal imagery, and this is also true of kinship terminologies. Quoting me, he notes that kinship terminologies always use spatial imagery. Such spatial categories appear to him to be innate, but their relative weight can vary. The order in which such categories dominate one another act on random linguistic (referential) variation to produce the patterns of meaning in any given terminology.
Jones' Figure 2 offers his comparison of English and Seneca. This represents only three generational levels: self, the one above, and the one below. Both terminologies define more positions than these, and more importantly both use categories that Jones's analysis cannot recognize because they are cultural and local.
In fact, terminologies can be represented comprehensively. This can be done at several levels of abstraction, beginning with the kind of diagram produced during field elicitation that I have called a “kinship map” (Leaf 1971) and carrying through algebraic reconstructions demonstrating their underlying generative rules. I have published the kinship map for English previously, contrasting it with Punjabi and addressing the same issues of spatial representation that are relevant here (Leaf 2006; 2009, p. 92). Read (1984; Read & Behrans 1994) has demonstrated its mathematical logic, and Read and Fischer have constructed a computer analysis that can regenerate it and any other system that is available on the Internet with the Kinship Analysis Expert System (Read 2006).
The English terminology has three important features that Jones's analysis obscures or misses entirely, and that OT apparently cannot encompass. First, positions can be extended vertically forever: Just keep adding the prefix “great.” The same number of prefixes mark the reciprocal terms. Jones recognizes generation, but not the idea of extensibility. Second, by contrast, in the horizontal direction “cousin” is a boundary. There is no relation beyond cousin. All terminologies must logically have boundaries in order to be closed, but Jones does not recognize the problem of finding them. And third, cousin in English is actually not specific to a single generation, so generation as a constraint does not dominate all others.
Every kinship terminology defines kinship in its own specific way based on its own conceptual contrasts. There are resemblances but no precise universals. The English idea of being related is strongly associated with the idea of common ancestry. To produce it, the terms consistently distinguish lineal versus collateral relation, sex, and generation in the direct descent line only. Outside the direct descent line, the terms distinguish only whether one is a sibling or sibling-spouse of a lineal or a descendant of such a sibling-spouse. The idea of an ancestor who is not a relative is a self-contradiction.
Seneca is indeed very different. Figure 1 is the kinship map for a female self (indicated by the gray circle). I have constructed it by asking the table of terms in Morgan's Ancient Society for definitions beginning with the core positions around self, just as I would have asked a group of live informants. Despite two terms that appear to be variant pronunciations of a single name, the consistencies emerge clearly.
Figure 1. Seneca, Female Speaking
Here, the +2 and -2 generations are boundary positions. The terms for the +3 and -3 generation are the same, although Seneca recognize individuals who are their more distant ancestors. So the second feature is that an ancestor is not necessarily a relation. The third conspicuous feature is that positions form groups based on a specific contrast between own matrilineal clan as against all other clans – not “matrikin.” Own brothers and sisters are grouped with mother's sister's children and contrasted with all other relations on one's own generation. Children of ha-nih (father and father's brother) who are also children of own mother are in own sibling group, children of other ha-nih are in the all others group. The same logic applies on the -1 generation. Finally, the difference between the male and female terminologies is that in the male terminology own children are grouped with other children of males on one's own generation, which Morgan does not explain but which seems to emphasize again the importance of the difference between descent through males and descent through females – as did Iroquois political organization. What the terms consistently distinguish are generation, clan, mother's and father's side, marital versus descent relation, sex of speaker, and sex of linking relative.
Seneca terminology embodies Seneca social conceptions just as English terminology embodies English social conceptions. Seeing how they do so brings us face to face with the cultural and social basis of thought itself. Empirically unnecessary speculation on the possibility of innate ideas is a distraction.
You have
Access
Jones assumes a referential theory of meaning drawn from ethnoscience and logical positivism. He cites Dwight Read and me without recognizing that we argue for a very different, pragmatic, conception (Leaf 1971; 2007; 2009; Read 2001; 2007). Pragmatic theory is demonstrably more parsimonious, precise, and powerful.
Referential theories are set up in terms of words and their referents. Pragmatic theories are set up in terms of ideas and their uses. Jones offers a list of universal parameters for classifying what kinship terms refer to and justifies it by invoking the supposed success of OT in linguistics. Read and I, by contrast, have demonstrated (not just suggested) that terminologies are systems of interrelated definitions, and that their internal order derives from computational rules contained in the core of terms that define the “direct” relations around the user, or self. Referring to relatives is just one of the many uses of such conceptual systems.
Jones begins by distinguishing two levels of language: “ordinary” and “VIP.” Ordinary is “open class or lexical,” which “includes most nouns and verbs.” He describes VIP as “closed-class” but otherwise only by example. The examples are grammatical markers. He puts kin terms in the VIP class. It is a crucial step in the argument, and it is unjustifiable. Languages have many closed classes, including classes of nouns and verbs. Kinship terminologies do form closed systems, but not classes in this sense.
Jones then argues that grammar uses spatial and temporal imagery, and this is also true of kinship terminologies. Quoting me, he notes that kinship terminologies always use spatial imagery. Such spatial categories appear to him to be innate, but their relative weight can vary. The order in which such categories dominate one another act on random linguistic (referential) variation to produce the patterns of meaning in any given terminology.
Jones' Figure 2 offers his comparison of English and Seneca. This represents only three generational levels: self, the one above, and the one below. Both terminologies define more positions than these, and more importantly both use categories that Jones's analysis cannot recognize because they are cultural and local.
In fact, terminologies can be represented comprehensively. This can be done at several levels of abstraction, beginning with the kind of diagram produced during field elicitation that I have called a “kinship map” (Leaf 1971) and carrying through algebraic reconstructions demonstrating their underlying generative rules. I have published the kinship map for English previously, contrasting it with Punjabi and addressing the same issues of spatial representation that are relevant here (Leaf 2006; 2009, p. 92). Read (1984; Read & Behrans 1994) has demonstrated its mathematical logic, and Read and Fischer have constructed a computer analysis that can regenerate it and any other system that is available on the Internet with the Kinship Analysis Expert System (Read 2006).
The English terminology has three important features that Jones's analysis obscures or misses entirely, and that OT apparently cannot encompass. First, positions can be extended vertically forever: Just keep adding the prefix “great.” The same number of prefixes mark the reciprocal terms. Jones recognizes generation, but not the idea of extensibility. Second, by contrast, in the horizontal direction “cousin” is a boundary. There is no relation beyond cousin. All terminologies must logically have boundaries in order to be closed, but Jones does not recognize the problem of finding them. And third, cousin in English is actually not specific to a single generation, so generation as a constraint does not dominate all others.
Every kinship terminology defines kinship in its own specific way based on its own conceptual contrasts. There are resemblances but no precise universals. The English idea of being related is strongly associated with the idea of common ancestry. To produce it, the terms consistently distinguish lineal versus collateral relation, sex, and generation in the direct descent line only. Outside the direct descent line, the terms distinguish only whether one is a sibling or sibling-spouse of a lineal or a descendant of such a sibling-spouse. The idea of an ancestor who is not a relative is a self-contradiction.
Seneca is indeed very different. Figure 1 is the kinship map for a female self (indicated by the gray circle). I have constructed it by asking the table of terms in Morgan's Ancient Society for definitions beginning with the core positions around self, just as I would have asked a group of live informants. Despite two terms that appear to be variant pronunciations of a single name, the consistencies emerge clearly.
Figure 1. Seneca, Female Speaking
Here, the +2 and -2 generations are boundary positions. The terms for the +3 and -3 generation are the same, although Seneca recognize individuals who are their more distant ancestors. So the second feature is that an ancestor is not necessarily a relation. The third conspicuous feature is that positions form groups based on a specific contrast between own matrilineal clan as against all other clans – not “matrikin.” Own brothers and sisters are grouped with mother's sister's children and contrasted with all other relations on one's own generation. Children of ha-nih (father and father's brother) who are also children of own mother are in own sibling group, children of other ha-nih are in the all others group. The same logic applies on the -1 generation. Finally, the difference between the male and female terminologies is that in the male terminology own children are grouped with other children of males on one's own generation, which Morgan does not explain but which seems to emphasize again the importance of the difference between descent through males and descent through females – as did Iroquois political organization. What the terms consistently distinguish are generation, clan, mother's and father's side, marital versus descent relation, sex of speaker, and sex of linking relative.
Seneca terminology embodies Seneca social conceptions just as English terminology embodies English social conceptions. Seeing how they do so brings us face to face with the cultural and social basis of thought itself. Empirically unnecessary speculation on the possibility of innate ideas is a distraction.