sign in sign up
<<
Home , Chiapas Zoque, Mayan languages, Nagual, Olmec religion, Oluta Popoluca, Sierra Popoluca

Society for American Archaeology

A Linguistic Look at the Author(s): and Source: American Antiquity, Vol. 41, No. 1 (Jan., 1976), pp. 80-89 Published by: Society for American Archaeology Stable URL: http://www.jstor.org/stable/279044 Accessed: 24/02/2010 18:09

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/action/showPublisher?publisherCode=sam.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Society for American Archaeology is collaborating with JSTOR to digitize, preserve and extend access to American Antiquity.

http://www.jstor.org 80 AMERICAN ANTIQUITY [Vol. 41, No. 1, 1976]

Palomino, Aquiles Smith, Augustus Ledyard, and Alfred V. Kidder 1972 Marriage patterns of the Chajul Ixil. Ph.D. 1951 Excavations at , . Carnegie dissertation, University of , Irvine. Institution of Washington Publication 594. 1972 Patrones matrimoniales entre los ixiles de Thompson, J. Eric S. Chajul. Guatemala Indigena, Vol. VII, Nos. 1-2. 1970 history and . University of Instituto Indigenista Nacional, Guatemala. Press, Norman, OK. Parsons, Elsie Clews Tozzer, Alfred M. 1936 Mitla, town of the . University of 1941 Landa's Relaci6n de las cosas de Yucatian. A Chicago Press, Chicago. translation. Edited with notes. Peabody Proskouriakoff, Tatiana Museum of Archaeology and Ethnology Papers, 1963 Historical data in the inscriptions of Vol. 18. Cambridge, MA. . Estudios de Cultura Maya 3:149-67. Villa Rojas, Alfonso 1945 The Maya of east central . Ruz Lhullier, A. Institution of Washington Publication 1954 en 1952. Anales Carnegie Exploraciones : 559. del I.N.A.H. 6:79-110. . 1947 Kinship and nagualism in a Tzeltal com- Sahagun, Fray Bernardino de munity, southeastern Mexico. American An- 1961 Florentine Codex: General history of the thropologist 49:578-88. things of , Book 10, The people, 1963 El nagualismo como recurso de control translated by Charles E. Dibble and Arthur J. social entre los grupos mayances de , 0. Anderson. School of American Research and Mexico. Estudios de Cultura Maya 3:243-60. the University of Publication 14, pt. 11. Villacorta Calderon, Jos6 Antonio Saler, Benson 1938 Prehistoria e historia antigua de Guatemala. 1964 Nagual, witch, and sorcerer in a Quiche Topografia Nacional, Guatemala. village. Ethnology, 305-28; reprinted in , Vogt, Evon Z. , and curing, edited by John Middle- 1969 Zinacantan, a Maya community in the ton. The Natural History Press, Garden City, highlands of Chiapas. Harvard University Press, NY. Cambridge, MA.

A LINGUISTIC LOOK AT THE OLMECS

LYLE CAMPBELL TERRENCE KAUFMAN

This paper explores the hypothesis that the archaeological Olmecs, at least in part, were speakers of Mixe-Zoquean . The hypothesis is supported by not only geographical and temporal correlation, but by Mixe-Zoquean loan words in other , many of which refer to things diagnostic of the Mesoamerican culture area. Also the cultural inventory revealed in Proto-Mixe-Zoquean vocabulary provides additional support.

A paper on Olmec might seem responds closely to that of the Olmec archae- pretentious, since presumably the last Olmec ological sites (Fig. 1, map of Olmec-MZ area), died long before any linguistic records were suggesting as a hypothesis for further investiga- made. However, the linguistic identification of tion that the archaeological Olmecs, at least in the Olmecs is a recurring question in anthropo- part, may have been speakers of Mixe-Zoquean logical literature (cf., for examples, Jim6nez- languages. To our knowledge, this hypothesis Moreno 1942; Coe 1968; Bernal 1969; Joesink- was first presented by Terrence Kaufman Mandeville 1972; Sharer 1974; and others). (1969a, 1973, 1974), who argued that the This interest, however, seems to have generated glottochronological time depth of MZ of 3,500 little more than poorly founded linguistic spec- years (around 1500 B.C.) correlates with the ulations, which would seem to justify a re- first glimmerings of Olmec civilization. examination of the linguistic identification of Although the geographical and temporal the Olmecs. The purpose of this paper is to correlation of MZ languages with Olmec civil- examine one particular hypothesis in depth, ization leads to sympathy for the Olmec-MZ that the Olmecs, at least in part, were speakers identification, the strongest support comes of Mixe-Zoquean languages. from purely linguistic considerations. We will The geographical distribution of speakers of consider first MZ words borrowed into other Mixe-Zoquean (henceforth MZ) languages cor- Mesoamerican languages, followed by implica- I - /

A CHALCATZINBO I FIGURE1. OLMEC-MIXE-ZOQUEAREA

k ZAPOTES _3 _ / % I / / .- _ % TSOTBAPAN(vcz)

\", " % If * OLUTA(VCM) lA I \ tt ' I * TEXISTEPECV \ * A \ I / SAN ? I t SAYULA(VCM) I !I LORENZO %.. I .- / / ' I / \ I TEAPA I '\ (TZ) / I / _I* .TAPU LAPA (TZ) \ VILLA '\ -% \ / ,C . PUXCATAN(TZ) OXOLOTAN I *ALTA(QM) CHOAPAN (OM) '. \4AGOALENA (CZf) o (TZ) TO)TONTEPEC (OM) m __ / * CHAPULTENANGO(CZ) -om * OCOTEPEC OCOTEPEC (CZ) S * (OM) SANJUAO 0 I I * COPAINALA (CZ) I AYUTLA (OM) UCHIOVI (OM) I SANTA -I CAMOTLAN(OM)* MARIA * oo * * CHIMALAPA(OZ) I USUMACINTA(CZ) / JUQUILA(OM) * COATLAN (OM) ,/p TUXTLA I SAN MIGUEL * * GUTIERREZ (CZ) I" CHIMALAPAIOZ/ OCOZOCAUTLA (CZ) *JUCHITAN e I *YAUTEPEC (OM) * TEHUANTEPEC W %

LANGUAGES 1

CZ= CHIAPASZOQUE OZ= OAXACAZOQUE VCZ= ZOQUE TZ= ZOQUE CM=CHIAPAS MIXE O M= MIXE

VCM=VERACUZ MIXE \ %MILES50 / 1 00 a \ sJM Isis

Fig. 1. Olmec-Mixe-Zoque Area. 82 AMERICAN ANTIQUITY [Vol. 41, No. 1, 1976]

tions of the reconstructedProto-Mixe-Zoquean struct only a limited PMZ vocabulary (about (PMZ)lexical items for this hypothesis. 450 items). We are relatively certain that as The Olmecs greatly influenced con- more Zoquean material becomes available we temporary groups and later cultures, and some will be able to expand the number of PMZ would claim that all the succeeding Mayan and lexical items. Therefore, items we list as Proto- Mexican cultures have their roots in Olmec Mixean (PMi) which are based on the Mixean civilization.The loan words from MZ into other languages may actually reflect PMZ items for Mesoamericanlanguages seem to reflect this which we as yet lack Zoquean information. extensive Olmec influence. Many of the loans refer to things which are diagnostic of the THE MIXE-ZOQUEAN LANGUAGES Mesoamerican culture area Kirchhoff (cf. Languages of the MZ family are spoken in If a culture must have such items to 1943). the southern Mexican states of Oaxaca, as Mesoamericanand the terms for the qualify Chiapas, Veracruz, and Tabasco. There are two items are borrowed from MZ, then it would main branches, the Mixe branch and the Zoque seem reasonable to assume that of the speakers branch. There are three main groups in the MZ the Meso- languages possessed uniquely branch. Chiapas (CZ) has sub- american and had Zoque Zoque things early enough prestige types: Central Zoque (including that others borrowedfrom them. If MZ Copainala), enough Northern Zoque (includingMagdalena), North- had the items and others lacked it would them, eastern (including Chapultenangoand seem reasonable to MZ with a culture Zoque equate Ocotepec), and Southern Zoque (including known to have had them at the appropriate Tuxtla Gutierrez and Ocozocuautla). Similarly, Like diffused of time, namely Olmec. aspects Veracruz (VZ) has subtypes: Sierra MZ loans extend Zoque Olmec material culture, many (including and about 25 into remote geographically very languages, other villages), and Texistepec Popoluca never known to have had a common frontier (spoken in Texistepec). Finally, Oaxaca Zoque with MZ (e.g., Xinca, , Jicaque, Paya, is the third main group (spoken in San the loan (OZ) etc.). Below we present word evidence, Miguel Chimalapa and Santa Maria Chimalapa). but first some comments about sources of our There is also Tabasco Zoque, apparently a the MZ and the identi- information, languages, divergent , but we have no information in order. fication of loan words are on the . The sources of our linguistic data are: PMZ There are also three main groups in the Mixe (Kaufman 1963), (our own branch. Veracruz Mixe (VM) has the two field notes, Kaufman 1964a, 1969b, Campbell subtypes of (spoken in n.d.a), Nahua (Molina 1571), (Asch- Sayula), and (spoken in Oluta). man 1962, 1973; Reid and Bishop 1974), Xinca Oaxaca Mixe (OM) also has two subtypes, one (Kaufman and Campbell n.d.), Lenca (Campbell with conservative vocalism (districts of n.d.b; Lehmann 1920), Jicaque (Campbell field Yautepec, Tehuantepec, and Juchitan, in- notes), Paya (Dennis Holt field notes), cluding such towns as Juquila, Camotlan, Cacaopera (Campbell 1975; Lehmann 1920), Coatlan, and San Juan Guichicovi), the other Matagalpa (Lehmann 1920), Miskito (Lehman with innovative vocalism (districts of Villalta 1920), Sumu (Lehmann 1920), Otomanguean and Choapam, including such towns as Toton- Lehmann languages(Rensch 1966;Pickett 1965, tepec and Ayutla). Finally, the third Mixe de Miranda Pride and 1920; Fernandez 1961; group is Chiapas Mixe (CM) or Tapachultec Pride 1970; Dyk and Stoudt 1965), Chontal (spoken only in Tapachula, probably now (Tequistlatec) (Turner 1971), Huave (Warkent- extinct). (Cf. Kaufman 1964b.) (The asterisk in and Warkentin 1952). we employ is standard linguistic notation to of these loans have been identified in Some indicate items of the proto language which are different contexts in our previous publications not attested but reconstructedusing the com- n.d.a, n.d.b; Kauf- (cf. Campbell 1972, 1975, parativemethod.) man 1969a, 1970, 1973, 1974, etc.). LOAN WORD CRITERIA Finally, it is important to point out that the very limited amount of available material on Questions about how one determines loan Zoquean languages has allowed us to recon- words and their direction need to be antici- REPORTS 83

pated. Our criteria for determiningloans and ?Iiw 'squash' has no cognate in other Mayan the direction of their borrowing are rather languages,so that PMi is the source (donor) and standard and well known in the linguistic Huastecthe borrower(receiver). literature. SemanticDomains Etymology (or MorphologicalComplexity) A weaker kind of inference comes from the The morphological composition of loan semantic domain of a loan. For words can give an indication of the directionof suspected example, in English things like squaw,papoose, borrowing. Typically in cases of borrowing wigwam, tomahawk, wampum, etc., with (barring unforeseen folk etymologies) the syn- onyms "Indian," "Indian donor languageis the one in which the item in involving i.e., woman," "Indian borrow- question can have an etymology which is baby," etc., suggest from Indian in Xinca morphologically complex while the receiving ing languages.Similarly most terms for can be shown to be language does not. For example, English cultigens aardvarkis seen to come from Afrikaansaard- borrowed from Mayanor MZ languages,so that other term to this vark, literally since the Afrikaans any belonging semantic "earth-pig," domain can be of form has a morphologicallycomplex etymology suspected being borrowed, and sources this is a while the English form is monomorphemic. possible sought. However, heuristicdevice and not a Similarly,PZo *nas-o?na?'fog' (literally "earth- proof. cloud") with its morphologically complex Phonology is seen to be the source from which etymology The strongest inferences are and possible using Tojolabal, Chuj, Jacaltec, Kanjobal, phonological criteria.Words that violate Motozintlec borrowedtheir forms typical ?asun,?ason, canonical forms are 2aso.'n'cloud'. (or morpheme structure) highly likely to be loans. English, for example, typically lacks initial clusters such as ?m, gn, Vl Geographicaland EcologicalClues (some speakers are incapable of pronouncing The geographicaland ecological remoteness these), so that words like schmaltz,shnook, and of gnu, impala, aardvark,cola, etc., make them schlemiel are nearly obvious loan words. Proto- likely candidates for loan words, and indeed Mayan typically had monosyllabic roots (with these were borrowed when English speakers very few disyllabic forms), and most roots in entered areas where they are found. Similarly, modern Mayan languagesare monosyllabic, so since cacao did not grow in the Uto-Aztecan that most polysyllabic forms can be suspected homeland (somewhere, one supposes, in north- of being loans or morphologicallycomplex. So western Mexico or the southwestern United tunuk'/tuluk' 'turkey' in Tzeltal, , Chuj, States), Nahua kakawa-'cacao' is likely to be a Jacaltec, and Motozintlec is a probable loan, loan (it is from PMZ *kakawa). Although and comparisonto PZo *tu ?nuk'turkey' proves inferences from geography and ecology are it to be so. Also (?)ko.ya. 2 'tomato' in often weak, the fact that cacao is thought to Cakchiquel, Tzutujil, , Aguacatec, and have been domesticatedin the very area where Chol is from PMZ *koya. For another kind of the MZ languages are spoken strengthens the example, Proto-Uto-Aztecan *p- was lost inferencein this case. initially in Nahua words, so any word contain- ing an initial p- in Nahua is to be Cognates likely borrowed. Nahua petla- 'woven mat' (petate) Whena suspectedloan occurs in many of the has the aberrantp-, it is from PMZ *pata. languages of one family with regular sound Obviously the best cases for identifying correspondencesso that it is reconstructablein loans and the direction of their borrowingare the proto language, but occurs only in one those in which a number of these criteria language(or a few languages)of another family, converge to leave little doubt. The Nahua case then the source is one of the languageswhich of "woven mat" is a good example. Nahua has cognates in its sister languages. For ex- pet(l)a- has no cognates in other Uto-Aztecan ample, Proto-Mixean(PMi) *(ti?wa 'squash'has languages (though some have borrowed this cognates in the Mixean languagesand is recon- term from Nahua quite late); there are cognates structable in the proto-language,but Huastec in MZ. Furthermore, it has the unexpected 84 AMERICAN ANTIQUITY [Vol. 41, No. 1, 1976] initial p- in Nahua. Another good example is Totonac manta; Jicaque mina; Xinca (?)ko:ya.2 'tomato' in a few Mayan languages, mula (perhaps);Chinantec mi3 1; Cuicatec which lacks cognates in the rest of the family. mi1 3. It also violates the typical monosyllabic root (8) Edible Tuber (chayote [huisquil], structureof Mayanlanguages. camote): PMZ *kdh-Zapotec gu; Chatino We have employed these criteriain consider- kuu 'sweet potato' (camote). ing the loan words presentedin this paper. (9) PMZ *sapani 'plantain'-Huave sapan 'zapotillo' (there are very few Huave forms with initial s-; mostly they are LOANWORDS Spanish loans). Plantains, though of re- Now we turn attention to the MZ loans cent introduction, were very often added found in other Mesoamericanlanguages. to the Zapote semantic domain in many Mesoamericanlanguages. BorrowedCultigens (10) Guava: (Zoque) pdtaq-- Tzeltal pdta, Tzotzil poto, and in several (1) Cacao: PMZ *kakawa-pan-Mayankakaw other Mayanlanguages (the native Proto- (Chol kdkaw, Tzotzil kokow) (violates form is The exceptional Mayan monosyllabic canonical Mayan *kaq'). typical first stress in Tzeltal and Tzotzil Nahua kakawa-(lacks cognates in form); show these forms to be loans. other Uto-Aztecan languages,not found Zoque ?olo-Xinca in the Uto-Aztecan homeland); Totonac (11) Papaya: Copainalf ueun, Nahua oeonih-tli, perhaps Ixcatec kakaw; Jicaque khaw; Paya kaku; Huave tYu2 u2. kakaw; Lenca kaw; Tarascan kahekua; PMZ *ka?wak kawak etc. (12) 'zapote'-Huave 'chico zapote'; ti-ka:?wa 'ciruela' (2) Gourd: PMZ *?ima-pan-Mayan ?ima the monosyllabic form, (plum? ). (not expected PMZ manioc accented on the first (13) *pisi (yuca)-Totonac pisisi sometimes syllable); also borrowed are: etc. guacamote. Perhaps Jicaquesem; PZo *?oha 'cotton'-Salvadorean Lenca (3) Squash (ayote): PMi */?i7wa-Huastec /'uwi 'cotton' (since cotton terms are /iw (lacks cognates in other Mayan lan- widely borrowed elsewhere from Mayan guages);Salvadorean Lenca ?/'iwan;Xinca and other languages);PMZ *tdapa 'greens' tii wa (perhaps);Tequistlatec (Chontal of (quelite)-Huave eapin 'tomato'; Xinca Oaxaca) -e?wa (a kind of squash, Dapa 'huisquil'(chayote). Calabaza de vichi); Ixcatec eu2; Chorti It is significant that such important Meso- e'iwan chayote (huisquil). american as beans, squash, tomatoes, (4) Squash, Gourd (calabaza):Oaxaca Zoque cultigens cacao, etc., were widely borrowed in ? awa-Tequistlatec -? awa; perhaps also gourds, Mesoamerican from MZ. It Miskito iwa, Honduran Lenca ewa, languages supports the MZ-Olmec since we can expect Matagalpaiwa, and Cacaoperaiwa. Com- hypothesis, others to borrow from the Olmecs as pare also Xinca #2wa 'squash' and cultigens the first civilized of Meso- Chontal -lewd2 'gourd'. We feel that highly agriculturalists america. of these cultigens are diagnostic probably Paya te2 wa 'chile pepper' is Many of and that such typically Meso- related also, since one species of squash , americansas and others should borrow has chile in its name, e.g., Ixcatec Mayans these terms attests the and powerful &u2-hina(literally &u2 'squash' and -hna' prestigious of MZ must have 'chile pepper'). position speakers languages had. (5) Tomato: PMZ *koya-Chol koya ; Maim, Aguacate, ?it-ko.ya2; Teco 9-ko.ya?; TheMaize Complex Cakchiquel,Tzutujil (?)ko:ya.: . Preparation (6) Bean: PMZ *sdk-Paya sak-. Most of the other languagesof the southernperiphery Terms involving maize and its preparation of Mesoamericahave borrowed the term for food are widely borrowedin Mesoamerican for beans from Mayanlanguages. languages.Though many of these were perhaps (7) Sweet Potato (camote): PMZ *manE- diffused widely at an earlier time than the REPORTS 85

cultigens discussed above, some seem to sup- (22) Axe (human ?): PMZ *pus to cut port the MZ-Olmechypothesis. with a knife or axe, *pusan metal (14) PMZ *way 'to grind corn', *waye (axe? )-Nahua pus-teki to cut, te-pos-(tli) 'pozole'-Proto-Mayan *wah tortilla; axe, metal (literally te- 'someone' plus Totonac wa? t tamal; Xinca iwa to make -pos 'cut', or "people cutter"); Pokom tortillas; Jicaque we nixtamal (leached pos stone war axe, ax pos 'wonder corn), tamal, we pirnmcorn dough; Proto- worker', pus 'witch' (encantador); Chiapanec-Mangue*wih? tortilla; Proto- Cakchiquel pos polished stone: Quiche Chinantecan*wih(n) tortilla;etc. pos, pus 'to sacrifice men by removing (15) PMZ *pi?i nixtamal (leached corn)- their hearts',to cut, polished stone, magic po.-?a tamal-Tzeltal, power; Cakchiqueland Quich6pus-nawal Tzotzil pal tamal; Xinca pa?''i to grind, 'magic power, witch'; Huaveapag to chop corn dough; Totonac pa? s(a) to shell with axe (hachear); Proto-Central corn; Nahua pa?a to grind, mash (per- Otomian *bes-na metal, lead; Proto- haps, Tequistlatec -gpa!'a?i tamal de Popolocan *pos hard stone. From these elote; Huavepeal tortilla. examples it seems certain that terms for (16) PMZ *po?t to grind corn, *po?te "axe" were widely borrowed from MZ, pinole-Nahua potonki harina muy and probably its significance was molida. borrowed as well. Both Caso and Bemal (17) To grind: PMZ *hod-Proto-Mayan*xue'; (reported in Bemal 1969) believe the Xinca hu?'i nixtamal,corn dough. Olmecs practiced , which (18) Maize: terms for maize itself are widely is also suggestedin these loans. Certainly, borrowed throughout Mesoamerica, in any case, the Olmec votive axes are though probably not from MZ. Ex- well known. amples: Proto-Mayan *?e?m; Tarascan (23) PMi *na?wa(y) old man ema; Xinca Lenca ama; (also husband)- ayma; ima, ema, Huave Nahua nawal and Sumu neawaneay witch; Cacaopera Matagalpaayma; witch, transformer,alter na-nawa-tia ama; Proto-Mixtecan Proto- ego, *yam; to transform oneself into an animal; Chiapanec-Mangue*-ma; PMZ *'k. Proto-Chiapanec-Mangue*nu-hwa witch; Perhapsalso borrowed is PMi *na(?)nto eat etc. The notions "(old) man" and atole-Mangue nambo atole. "witch" are related in languagesthrough- Ritual and CalendricTerms out Mesoamerica,e.g., in many Otoman- guean languages(cf. Rensch Xinca (19) Incense (copal): PMZ *po'mV-general- 1966). borrowed the word for "man" Mayan po'm (but Huastec Xinca Mayan hom); with the meaning "witch." The Nahua pu.mu; Totonac Tepehua pu-m; pu-m; ( and call them- Tequistlatec -boma; Huave pom. Incense Pipil speakers) selves nawa 'people', the man- seems always to have been indispensable showing to Mesoamericanritual. witch association even within Nahua. Though nagualismis widespread,it seems (20) PMZ *may 'to count, to divine'--Kekchi not unlikely that the term bor- may twenty, twenty years; Pokom may widely rowed in Mesoamerican came twenty years; Quiche and Cakchiquel languages from the Olmecs. may twenty years of 400 days each, may q'i.x the calendar; perhaps also Kekchi (24) Woven Mat (petate): PMZ *pata-Nahua mayex sacrifice, offering; mai- pet(l)a-tl (from Proto-Aztec *pata); measure(ment). Proto-Otopamean*pe. The petate is well (21) PMZ *?ukA, PMi *9ok dog-Yucatec ok known as a symbol of both secular and 'dog' calendric day name: Huastec ok religious rank and power in Mesoamerica. fox. The Kanjobalan group of Mayan Aztec rulers and and Eagle war- languages has ?o?q, ?oq '', which riors were seated on them, as were the be a may possible cognate of the Huastec Quich6 leaders (hence the name of the word, although we think the Yucatec "Book of the Council" from form is a true borrowingfrom MZ. po'p-o'l 'council',from "mat").The direc- 86 AMERICAN ANTIQUITY [Vol. 41, No. 1, 19761

tion from MZ into Nahua is quite clear. Chatino *su;etc. Nahua's pet(l)a-tl lacks cognates in other (33) Perhaps PMZ *sam to heat something- Uto-Aztecan languages (though some WesternMayan *sa2 m(-et) griddle. Mexican Uto-Aztecan languageshave bor- (34) Perhaps CopainalaZoque ko-pa?e trap- rowed the term from Nahua). Proto-Uto- WesternMayan *peh?'. Aztecan initial *p- was lost in Nahua, so OtherLoans that this form with its initial p- is (35) PMZ *?/ukmouse-Chol luk. (The Proto- anomalous. Finally, the Pochutec form is Mayanword is *'o 2 h.) pot, where the o (correspondingto other (36) PMZ *aa'm ripe, good-Tzeltal ?am good; Nahua e) reflects Proto-Aztecan *d, mak- Xinca ?'ama good. ing Proto-Aztecan *pata much more like (37) PMi *wa.s fox-some Mayan languages PMZ *pata, from which it was borrowed. have wa ?; Tarascanxiwa#ei (25) Paper: PZo *toto-Mixtec tutui. Perhaps (38) PZo *we tu fox-Tzotzil, Yucatec, also, PMi *nokE-Huave nawi g. Tojolabal, Jacaltec, Mam *we.t; Xinca (26) Turkey: Pzo *tu?nuk-Tzeltal, Tzotzil, we-to; Mixtec (vidzu) (calendric day Chuj, Jacaltec, and Motozintlec tunuk'l name) fox. tuluk'. This violates the typical Mayan (39) PZo *nas-o?na fog (from *nas 'earth' + monosyllabic root structure (the native *?o na 'cloud')-Tojolabal ?ason cloud; Mayan form is *?ak'). The PMi form Motozintlec ?aso n cloud; Kanjobal, Jac- *tu.tuk (and *tu't to lay eggs) together altec, Chuj ?asuncloud. with the PZo form, is probablyrelated to (40) PZo *?une PMi *?unak child-Tzeltal, Tequistlatec -dulu turkey; Jicaque tolo; Tzotzil, Tojolabal, Chuj, Kanojabal, Huave tel female turkey; Zapotec tou? Jacaltec, Cholti, Mam une/unin; Xinca turkey; Nahua totol- chicken (toto-tl ?one child, immature;Otomi uene baby; bird); and Paya totoni- chicken. Since perhaps also Nahua kone-tl child (which domesticated turkeys appearquite late in lacks cognates in other Uto-Aztecan lan- Mesoamerica(around A.D. 300; Michael guages). Perhaps the importance of in- Coe personal communication), it is not fants in Olmec art motifs, and therefore certain how these forms are to be inter- presumably also in Olmec religion, con- preted, perhapsas later loans. tributed to the wide-spreadborrowing of (27) Bee, Wasp, Wasp's Nest: PMZ *2a.kaw, this term from MZlanguages. Sierra Popoluca (Zoque) ?okwor- (41) PMZ *u.ma deafmute-Chol, Tzeltal, Huastec 2okow; Tzeltal, Tzotzil, Tojo- Motozintlec ?uma (the native Proto- labal 2dko(with unexpected first syllable Mayan word is *me.n or *me.m). stress);perhaps also Mixtecyoko. (42) PMZ *pa?i lizard-Cakchiquel, Quich6 (28) Sandals: PMZ *ke?ak-Nahua kak-(tli); (?)pa2&, Pokom patis; Tequistlatec compare this to Proto-Otomanguean -bati ?. This may have been borrowed as **(h)kWa(h)(n)2 (which may not be a calendricterm, comparethe Otomi day based on real cognates); Proto-Popolocan name am-beflga lizard. *ka2; Proto-Chiapanec-Mangue*hkah? ; (43) PMZ *2uspi(n) alligator-Totonac u0pi, the other Uto-Aztecan languages which u?~upi; Tepehua htukspi;Tarascan uspi. have borrowed this form are: Varohio This also may have been a calendricterm. kahkrd; Cora ka?akai; and (44) PMZ *koya rabbit-Huastec koy; Huave ka .kdi. koy. Perhaps also comparable are (29) PZo *?oH pulque, maguey-Nahua ok- Totonac skaw; and Otomi khwa; Mat- (tli) pulque. lazinca kwha. These are also day names in (30) Perhaps PMZ *ka'na salt--Huave kiniak. Otomi and Mixe, and probably in the (31) Perhaps Woven Mat (petate): PMi others as well, which may account for *to?kE-Huave tek; Totonac 9-ti0kat. why they were borrowed. (32) Perhaps Sayula Popoluca (Mixe) suy pot (45) PMZ *(hah)/uku? ant-Mixtec o6k6, (olla); Sierra Popoluca (Zoque) su? iy tiyoko; Nahua ?ika-(tl) (from Proto- pot-Jicaque soy; Pipil Aguh-;Nahua sok-; Aztecan *tikV, from earlier *($uk-V-); Xinca suh-; Proto-Otopamean*su; Proto- Huaveeok; Cacaoperasuku-l; etc. REPORTS 87

(46) Perhaps PZo *4i(? ) opossum (tacuazin, hypothesis to entertain it as a plausible,if not tlacuache)-Xinca se2, c'a'pa; Nahua yet provable, hypothesis. On the basis of this si? (-tli) liebre (jackrabbit?); Salvadorean assumption, we now turn to the potential Lencase-suli; Honduran Lenca sewe. contribution of comparativelinguistics to the (47) Perhaps PMZ *pa-hu? coyote-Paya understanding of Olmec culture. We will in- pa-ku ? ; Mixtec wd? U. vestigate the reconstructedPMZ vocabulary for (48) Perhaps PMZ *pok(A) gourd-Chol pok'; cultural content. In the interest of space we do Totonac po 'qo 2tnu ?; etc. not discuss the method in detail (for other (49) Perhaps PMZ *?afi elder brother- examples of this approachto cultureprehistory Quichean *?al; Mamean *2alik; Taras- see Longacre and Millon 1961 and McQuown can a4'-iwoman's younger brother. 1964). However, briefly stated, when firm (50) The Mixe day name (Juun)corresponding linguistic data are sufficient to reconstruct an to iguana, is probably the source of these etymon in the proto language, one usually loans (which are not cognates) in Mayan assumes (barring undiscovered complications) languages meaning iguana: Mam ?o n; that the referent of the etymon was part of the Teco xo ?on; Motocintlec ?o -ha ?n; cultural inventory of the speakersof the proto Quich6 ?o?on; Yucatec huh; Cholti (hu); language. Since this study shows that the PMZ Chorti hu(h); (cf. also Aguacatec and speakers had a rather sophisticated Meso- Ixil). american culture around 1500 B.C., this in- These loan words provide rather strong formation can be viewed as an additional support for the hypothesis that the Olmecs (at argument for the Olmec-MZ hypothesis. least in part) spoke MZ languages.The fact that Furthermore,one can speculate, in accordance many of these are in geographically quite with our hypothesis, that the culturalinventory remote languages shows their importance. The of PMZ was also part of the culturalinventory number and extent of these borrowingssuggest of the Olmecs: the same. Because so many are so central to Agriculture: *kama(?/n) milpa, *yu-h to everything Mesoamerican, and because MZ clear land, *ni-p to sow (plant), *tam seed, seems always to be giving but very rarely fruit, *puh seed, *?ik to harvest. receivingthese early loans, it does not seem to The Maize Complex: *mo-k maize, *way to be overstating the case to conclude that the grind corn, *pi?i leached corn (nixtamal), Olmecsprobably spoke MZ languages. *h -pak corncob, *waye posole, *po?te pinole, Of course in a brief paper with the goal of *po?t to grind (grains), *?9ks to shell corn, presenting a hypothesis to be tested in further *ham lime. research one cannot anticipate all possible Other Food Plants: *ni.wi chile peppers, objections, or even raise all the important *sak beans, *koya tomato, *manEsweet potato questions. For those who would like to see (camote), *pisi manioc (yuca), *kah edible these as mutual loans from some other un- tuber (chayote, camote); *nuhpe(n) choke- designatedlanguage into MZ and the languages cherry, *(y)a-ti(n) custard-apple (anona), listed here, it is important to point out that on *?owi? avocado, *ka?wakzapote, *kumacoyol the whole these fit MZ phonology and palm, *po?os ?igva, *kakawa cacao, *?opa canonical patterns with no difficulty, but not greens (quelite). those of any other Mesoamericanlanguage or Important Animals: *ma?a deer, *na-4 we know of. Furthermore, armadillo, *tao? iguana, *koya rabbit, *liku there seems little reason to seek phantom coati, *?i.nu honey, *?awi9 monkey, *kahaw languages, since even without the loan word jaguar. evidence, the matching location of MZ lan- Fishing: *Jak fish, *9e-si crab, *ma.k to guages and Olmec culture in time and space fish, *suy to fish with hook and line, *?a9 would suggest that MZ languagesare the best canoe. candidatel Textiles, etc.: *pit to spin thread, *nawin IMPLICATIONSOF PMZVOCABULARY agave (maguey) fiber, *taps to twist rope, *kah?ay hammock, ?ay cord, vine. We assume that the loan word evidence Ritual, etc.: *ma.san holy, *po-mV copal presentedis sufficient support of the MZ-Olmec incense, *pus to cut with knife or axe, *pusan 88 AMERICAN ANTIQUITY [Vol. 41, No. 1, 1976] metal (aboriginally axe?), *ha.y to write, *may origins or geographical extensions. It makes no to count, to divine, *9ev to dance, *kow to difference to our hypothesis if the Olmec play a musical instrument, *kowa drum, *saw origins are found in the heartland (Veracruz, festival (fiesta), *9ame year, *9ips twenty, Tabasco), the Pacific slopes of or *mo9ne bundle of 400, *?owi tobacco, *hu-kV Oaxaca, or the Morelos or Northeastern Oaxaca cigarette, hu -k to smoke. area (cf. Wicke 1971). The only thing crucial to Commerce: *to9k (Mi) to sell something, our hypothesis is that the Olmecs did occupy (Zo) to spread something out, *yoh to pay for, the heartland in the period we are talking *?ow to cost, be worth, *huy to buy some- about, and that does not seem to be very thing, etc. controversial. Furthermore, the distribution of Other: *Coy liquor, remedy, *pok(A) water MZ languages would seem to allow for any of gourd, *?ima gourd ('icara), *te9n ladder, *tak these hypotheses of Olmec origins without house, *kom house pole (horc6n), *me?esi difficulty. adobe wall, *na?a rubber, chicle, *ka?-kuma We conclude that probably the Olmecs, at ring, *top to shoot an arrow, *?e9es bed, *?e? least in part, spoke MZ languages. We hope this to plane wood, *kaoak sandals. hypothesis with our supporting evidence will the From the reconstructed lexical items of PMZ generate further research on topic. it seems that the speakers of PMZ practiced Aschmann, H. slash and burn agriculture (milpa, to clear land, 1962 Vocabulario Totonaco de la Sierra. Serie de and had a full complement of Meso- Vocabularios Indigenas 'MarianoSilva y Aceves' etc.), 7. Instituto de Mexico. and its Lingiiistico Verano, american cultigens (maize preparation 1973 Diccionario Totonaco de . Serie de complex, chile, beans, tomatoes, several root Vocabularios y Diccionarios Indigenas 16. In- crops, many fruits, and gourds). Fishing was stituto Lingiiistico de Verano, Mexico. important. They had textiles. If the MZ-Olmec Bernal, I. 1969 The Olmec world. University of California is then this linguistic evidence hypothesis true, Press, Los Angeles. confirms the archaeological evidence of these Campbell, L. things (cf. Bernal 1969). Furthermore, it seems 1972 Mayan loan words in Xinca. International that PMZ had already developed the vigesimal Journal of American Linguistics 38:187-90. 1975 system (twenty, bundle of 400). They Cacaopera. Anthropological Linguistics 17(4): 146-53. also must have had divination, nagualism, and n.d.a Quichean linguistic prehistory. University of some form of writing, among other things. California Press, Los Angeles. (In press, ms. 1975.) n.d.b last Journal CONCLUSIONS The Lenca. International of American Linguistics. (In press, ms. 1975.) Because the geography and chronology of Coe, M. PMZ and the Olmec correspond closely, we 1968 America's first civilization: discovering the Olmec. American Heritage, New York. suggested MZ languages as the most probable Dyk, A., and B. Stoudt candidate for the linguistic identification of the 1965 Vocabulario Mixteco. Serie de Vocabularios Olmecs. We considered loan forms from MZ Indigenas 'Mario Silva y Aceves' 12. Instituto into other Mesoamerican languages. Because Lingiistico de Verano, Mexico. M. of these loans refer to items of Fernandez de Miranda, T. many diagnostic 1961 Diccionario Ixcateco. Instituto Nacional de Mesoamerican culture, and some occur in geo- Antropologia e Historia, Mexico. graphically quite remote languages, we find that Jimenez Moreno, W. the MZ-Olmec hypothesis has considerable sup- 1942 El enigma de los Olmecas. Cuadernos Furthermore, the reconstructed PMZ A mericanos 1:113-45. port. L. items of cultural content a Joesink-Mandeville, vocabulary suggest 1972 Concerning Olmec-Maya relationships: a rather sophisticated Mesoamerican culture for correlation of linguistic evidence with archae- speakers of PMZ around 1500 B.C., additional ological ceramics. Paper presented at the 1972 support for the hypothesis. ASS meetings. T. Finally, it is important to point out that our Kaufman, 1963 Mixe-Zoque diachronic studies. Un- hypothesis that at least some Olmecs spoke MZ published manuscript. languages is in no way dependent upon any 1964a Materiales linguisticos para el estudio de particular solution to the questions of Olmec las relaciones internas y externas de la familia REPORTS 89

de idiomas Mayanos. In Desarrollo Cultural de Millon, R., and R. Longacre los Mayas, edited by E. Vogt, pp. 81-136. 1961 Proto-Mixtecan and Proto-Amuzgo- Special publication of the Seminario de Cultura Mixtecan vocabularies: a preliminary cultural Maya, Mexico. analysis. Anthropological Linguistics 3(4): 1-44. 1964b Mixe-Zoque subgroups and the position of Molina, Alonso de Tapachulteco. International Congress of 1571 Vocabulario de la lengua Mexicana y Castel- Americanists 35:403-11. lana. Mexico. 1969a Some recent hypotheses on Mayan diversi- Pickett, T. fication. Language Behavior Research Lab, 1965 Vocabulario Zapoteco del Istmo. Serie de WorkingPaper 26. Berkeley. vocabularios indigenas 'Mariano Silva y Aceves' 1969b Teco-a new Mayan language. Interna- 17. Instituto Linguistico de Verano. Mexico. tional Journal of American Linguistics Pride, L., and K. Pride 35:154-74. 1970 Vocabulario Chatino. Serie de vocabularios 1970 Precolumbian borrowings in and out of indigenas 'Mariano Silva y A ceves' 15. Instituto Huastec. Paper presented at the 6th annual Linguistico de Verano, Mexico. meeting of the Chicago Linguistics Society. Reid, A., and R. Bishop 1973 Areal linguistics and Middle America. In 1974 Diccionario Totonaco de de Current trends in linguistics, Vol. 11, edited by Juarez, . Serie de vocabularios y dic- T. Sebeok, pp. 459-84. Mouton, The Hague. cionarios indigenas 'MarianoSilva y A ceves' 17. 1974 Mesoamerican Indian Ency- languages. Instituto Linguistico de Verano, Mexico. clopaedia Britannica, 1974 edition. Kaufman, T., and L. Campbell Rensch, C. n.d. Xinca and dictionary. Unpublished 1966 Comparative Otomanguean phonology. Un- manuscript. published Ph.D. dissertation. Department of of Kirchhoff, P. Linguistics, University Pennsylvania. R. 1943 Mesoamerica: its geographical limits, ethnic Sharer, 1974 The of the southeastern composition and cultural characteristics. In prehistory Maya periphery. Current 15:165-87. Ancient Mesoamerica: selected readings, edited Anthropology P. by J. Graham, pp. 1-14. Peek, Palo Alto. Turner, 1971 Chontal to Lehmann, W. Spanish-English Dictionary, to of 1920 Zentral-Amerika. Berlin. Spanish Chontal. University Press, Tucson. McQuown, N. Warkentin, M., and C. Warkentin 1964 Los origenes y la diferenciaci6n de los 1952 Vocabulario Huave. Instituto de Mayas segun se infiere del estudio comparativo Linguistico Mexico. de las lenguas Mayanas. In Desarrollo cultural Verano, de los Mayas, edited by E. Vogt, pp. 49-80. Wicke, C. Special publication of the Seminario de Cultura 1971 Olmec: an early art style of precolumbian Maya, Mexico. Mexico. University of Arizona Press, Tucson.

ARCHAEOLOGICAL DATA BANKS IN THEORY AND PRACTICE

SANDRA SCHOLTZ G. CHENHALL

Although archaeologists have experimented with computers since the early 1960s, with a few exceptions, the concept of an archaeological data bank has not been readily accepted. The authors believe that data banks can be usable tools, but that they will be used only if they are designed to satisfy realistic and precisely defined needs, and only if adequate consideration is given to data structures, human problems, and theoretical issues. Just finding the right computer system is not enough.

The objectives of this paper are: (1) to data banks can be usable, but we have learned briefly trace some of the authors' experiences from experience that this does not happen just in ten years of data banking; (2) to report on an by finding the right "black box" computer archaeological data bank project presently system and then recording a large number of being carried on by the Archeological field or laboratory observations. Data banks Survey; and (3) to delimit both the possibilities must be created to satisfy realistic and precisely and the limitations that appear to be inherent defined needs, and they must be implemented in the storage and retrieval of archaeological with adequate consideration for the theoretical, data on computers. We strongly believe that human, and data structuring problems.