SETTLEMENT ARCHAEOLOGY Edited and with an Introduction by Yale University National Press Books (2) Palo Alto, CaliforniaXI GORDON R. WILLEY Settlement Archaeology: An Appraisal This volume is a collection of essays on various substantive and reflective aspects of the settlement pattern in archacology. The first two papers in the collection go directly to 2 controversy that has Tong been latent in our dis: cipline~a debate that has been expressed, if at all, only occasionally or obliquely. The gist of the controversy may be stated as 2 question: is there “settlement pattern approach” to archaeology or prehistory that may be said to constitute a “new archaeology," a radical break with the past of the discipline? A dozen or so years ago, when T organized a symposium on "Pre- historie Settlement Patterns in the New World,” 1 had not considered such a question, and 50 was surprised at the reaction of some members of the sym- pposium audience, in indicating that they believed such an issue to be at stake, These reactions prompted me to review my owa thinking on the ter. In my only previous writing on the subject of prehistoric settlement patteras (Willey, 1953a), I had presented my research as part of a larger pro: ject or “team” effort on the archecology of the Vira Valley of Peru. In this context, I relied to a great extent on other lines of archaeological investiga- tion-site survey and stratigraphic excavation, ceramic analysis, horizontal seriation, and the formation of culture complexes, phases, and periods by these means. I was at pains to point this out (op. cit. pp. xvii-xix, 9-12), but 1 also emphasized the nature of my own work as something of a new depar- ture in the“. . . understanding of the structure and function of ancient societies.” In considering this in retrospect, 1 concluded that the investiga- tion of settlement patterns did not, and could not, in itself compose 2 self- contained approach to prehistory, that it was not a “new archaeology.” At the same Lime, it was, to 2 very great extent, a new" approgch within ar chaeology, or” at least one relatively litle exploited by archaeologists, These views were stated in a brief introductory note to the published papers ‘of the aboye-mentioned symposium on prehistoric settlement patterns (Willey, 1956a} I adbere to these views today, and have expressed them again in a recent monograph (Willey, Bullard, Glass, and Gifford, 1965, pp. 5-6, $81). The reader will, I trust, forgive this digression into personal his- tory, for it has a bearing on the Chang-Rouse debate and on my own stti- tude toward that debate, 208 ee ty tes Settlement Archaeology: An Appraisal /209 ‘CHANG AND ROUSE: A QUESTION OF PRECEDENCE AND ExPuAsis ‘The differences of opinion between Chang and Rouse are largely pre ‘edural, and involve the sequence of steps in archacological investigation. ‘Chang argues that no part of archaeology or prehistory takes logical pre- cedence over any other part, that the interpretation of artifacts is not on a more basic and more reliable level than, say, the interpretation of religion Accordingly, the selection or acquisition of data is not a first step, because the very matter of choice in the selecting is conditioned by interpretive hypotheses. Rouse, on the other hand, takes the position that the first four logical steps of prehistory are: (I) recovery of the remains; (2) classification of the remains; (3) reconstruction from the remains; and (4) interpretation of the remains. It is his feeling that Chang, in his desire to proceed directly to the study of prehistoric society, omits the first ofthese steps. It seems to me that Chang does not so much ignore this step as he does tke it for granted, For procedurally it is impossible to define a meaningful prehistoric settlement pattern unless one begins with the unit of a cultural ‘complex; and cultural complexes are defined by the clusterings or associ tions of artifacts and remains at given points in space and time. Once such complexes are Known, through the procedures of recovery and classifica- tion, it is then possible to correlate settlement features into patterns that represent, or are the residues of, former social institutions, Before this is done, the individual settlement features are no more 10 us than bumps or marks upon the landscape-lacking in cultural identification, chronological position, or functional significance. Once these kinds of information are known, in any particular region or area, it may then be possible to evaluate 4 prehistoric settlement pattern on its formal settlement properties alone; ‘bat this is a secondary operation, underpinned by the primary procedures ‘of culture complex definition, Icis in a procedural sense that tools or artifacts are on a lower or more ‘asic level of archaeological interpretation than religion. This is quite dif- ferent from claiming that“ . .. the study of technology is both precondi- tional to and more reliable than theology.” The question here is not one of cultural determinism, or of the determinative influences of one aspect of ‘culture upon another, but of what must, of necessity, come first i field investigation, As archaeologists, we address ourselves first to the “hard goode"—sherds and flint and buildings, or whatever-because of the very nature of the discipline. Thus 1 would have to enter a disagreement, or ‘rong qualification, to Chang's statement that: “Studies of artifacts and of religions belong to different fields of inquiry...” They must, think, at the outset, represent no more than different aspects of a common inquiry, while the limits of form, space, and time are being staked out, After these formal-spatial-temporal systematics have been executed, a number of lines of inquiry will separate out,210/ GORDON R. WILEY Rouse is correct, then, I think, in holding that archaeologists have a sort of core discipline in common: the technical recovery, description, and classification of remains. He explains this further by stipulating that an ar- chacologist, so defined, must be more than simply a technician. He must be versed in theory—"". | the theory of material culture, especially of artifacts and structures.” I am not at all sure that { know what these “theories” are, or just what Rouse has in mind, In general, I should think, the methodolog- ical core of archaeological procedure is accommodated to’ the anthropolog- ical premises that culture, of cultures, change through time and that cultural ‘and social behavior is in some way registered in material objects. As archae- ‘ologists we deal with this behavior through the surrogate of these objec, arriving at conclusions about time and place and change, and building these conclusions with the aid of concepts such as diffusion, evolution, and eco- logical adjustment drawn from anthropology and other disciplines. With this outlook the archacologist has operated with certain rather well-defined techniques and with a growing battery of technical aids. Assuming that this foliows Rowse’s meaniag, 1 concede it to be the common ground of a chacology and understand him when he remarks that to exceed this in i terpretation carries us into the specialized realms of substantive knowledge: Egyptology, classics, or other culturally and regionally defined spheres of prehistory or art history. But Tam not at all content with Rouse's further insistence that archaeology (methods and techniques, with a body of guid- ing theory) should as a discipline stand apart from, say, prebistory (the substantive results of archacological recovery and thelr further interpreta- tion). Rouse develops this last idea at some length, and deseribes what he be- lieves t0 be a trend in disciplinary organization, The archaeologist of the future, according to his view, will have been brought up in a different howse from the prehistorian, He will be primarily the technician, the specialist in the retrieval of the data of the past, whatever that past, He and others like will compose a sort of “service” cadte to those who are concerned with the substance of human history. The latter, the prehistorians, will be the thinkers, the formulators of ideas. Although ackaowledging that we are in the age of the specialist, I look with dismay on the reduction of the archae- ologist to such a fragmentary man, or on the sterilization of the prehisto- rian by removing him from the dirt, sweat, and tears of his methods and techniques. One of the great strengths of American archaeology is that it has grown up in the merging currents of anthropology and technical archae- ology, As a result, the American archaeologist has been able to construct more sophisticated models of the past and to phrase a great range of new problems that would never have occurred to him if his outlook had been ‘one of technical and mechanical proficiency alone. On the other hand, the technically trained archacologist has been able to bring to anthropology a greater comprehension of the kinds of questions that may be asked of the nh eee ne et Settlement Archaeology: An Appraisal /2\1 ata of prehistory with reasonable hope that answers will be forthcoming, Rather than to separate archaeology from prehistory—or method from sub- stantive result I should prefer to see them continue as yokefellows in a fingle discipline, and, whenever possible, and according to the dictates of personal taste, to continue as aspects of the same anthropologically trained personality. In all of this, Iam closer to Chang's opinion, that archaeology, even as a field operation, cannot be separated fully from interpretation: there must be a coordinated problem direction to prehistory and to the ar- hacological methods devised to recover prehistory. This would certainly be trae of methods devised to plot and reconstruct prehistoric settlement pattorns—with overall problem direction in this case being, for example, the understanding of past social and political organization. My difference with Chang, as I have tried to make clear above, is not over the inevitability of & close relationship between archaeological field method and culture~ Fistorical interpretation, but over what I believe (along with Rouse) to be the necessity of taking certain basic methodological steps prior to the set tlement pattern-prehistoric society interpretive step. Procedural steps are, again, at the heart of the Chang-Rouse disagreement ‘over the meaning and interpretation of such concepts as “settlement,” “community,” and “components.” Chang says: “An archacological settle- °. ment i the’ physical locale or cluster of locales where the members of & community lived, ensured theit subsistence, and pursued their social func- tions in 2 delineable time period.” Such is for him the basic unit of the science of prehistoric society. Rouse recognizes the value of such a unit in paleosocial interpretations but questions its suitability as the fundamen tal building block. He argues that the settlement can be recognized archac- ‘logically only as “all those space/time components in which a particular “= community carried out its various activities,” not as a single component. 1 must admit that, heretofore, I have always considered the “component” in archaeological culture classification to be the equivalent of the social category “community” and of the physical entily “settlement” (Willey ‘and Phillips, 1958, p. 49). In other words, my thinking here has gone along with what Chang is suggesting. Rouse demurs, however, on procedural rounds, and asks: What is the primary unit perceived by the archaeolo- gist? Is it (in West Indian prehistory) the village site and the ceremonial ball- court, when these are found some distance apart? Or are the village site and the ball-court each components at the level of basic archaeological classification? It can be, and has been, demonstrated, in this particular in- Stance, that villages and ball-courts are remains left by the people of the fame community, They represent wo activities or aspects of the life of a community: domestic functions and ceremonialeecreational functions ‘Yet on initial investigation this may not be known. Should they, thea, be dlassified apart? And is this sort of stricture what the device of the “com- ponent” has implied and should imply? 1 would agree that such physically fartomeses det of aduies eekApotts ter prrck wert Comprntal 212/aoeo0w ». wittsy isolated entities should be considered apart in the eatly stages of ines tation, but T arm hesitant to class them as separate “components.” How, operationally, are we to. draw consistent lines? Are ballcourts, burial mounds, temples, or sweat-baths alvays to be considered as component apart from domestic structures? Or only when found spatially apart fom dwelling sites? And, if so, how far apart? 1. has not been general American archaeological practice to so. proceed, ither in those parts of the United States where the “component” concept has beon formalized as an clemeat of the Midwestem Taxonomic Sysiem or elsewhere; and I should think it mistake to do so. tre ~ (els watud @S om iabay Thus, while recognizing the validity of Rouse's argument thatthe con expt of is not on a primary level of investigation, 1 think the concept important on the secondary level of investigation-in- terpretation. Chang's definition ofthe seulement as the locus of a comms. nity i apt and useful; and I should like 19 equate settlement and community with “component.” I donot like the archacolopical concepts of either oeus" or “phase” asthe equivalent of settlement of community—as Rouse has suggested for focus and phase have an established usage that carries them well beyond the usual geographical, chronological, and typological dimensions we are accustomed o associate with a community My understanding of Chang's “objective types” and “relative types that of a dichotomy between more or less universal traits or trait categories, fn the one hand, and stylistically characterized traits or units, on the othe In this frame of reference, “objective types” are somewhat more than cri- ralogy or geometry. They are functional But that Louis XV chaie isa “celative type.” The first hes meaning inthe realm of general comparative inference and analogy, the second takes is tmcaning from stylistic identification ‘within a specific historical tradilion The terminology” (objctive” “relative") that Chang. has chosen here seems somewhat abstrse. As T understand it the qualities he Is aking bout ae those of function and ase, and a general comparative appreciation of thes, and, set agaist these, syle and historical uniqueness. The dies sions of “function” and “sty,” by both Chang and Rouse, relate to this fame question; and here [ am in agreement with both and with, Rous’ Statement hase and faneton abot served separately in ypoloy. In “microstructure” and “mactésieSiey "Chit offers useful and _ssicatc cones for the intasetement 0 Intresonmniypovne, an the one band, and the larger extraseltement or exira-community do: main, on the other. Rouse’s reactions are lirgly favorable but, again, tempered by the reservation thatthe systematics of archacologial modes, types, and phase building be interposed as basic to settlement considers tions, Here. atone point, Tam uncertain what Rovse intends. With refer. ence to the defining of macrosirucures, he remarks that “secements a © tein of desertion fom J. ‘Gn foe rout my tata ai on woe set heen, % wuhiafiginy “e ae | | Settlement Archaeology: An Appraisal /213 unsuitable, because they are formed by infering communities from the teang"'BUL i not aay archaclogca ype formed by making inferences from the remains? The poterysype~with is mods of shape, decoration, fem Mv at ith ever ftecer than the teen. pattern but ie ico ST omsthieas ioe. Tf we nso tht setdemont patern i 8 eee atetnd can, Ihe other ta be claified imo types this cse Sate boss of moves of dwellings, other Buildings, satel arrangements, Ste wty canna setfemen serve aya device inthe dition ofa ulare [Sapo ot cure phase, tnd inthe bling of macosroctural mols of Qiu and soley? Why cannot the "smal ceremonial coter-seatteed Fat setlomest pater be a eiteion of Pave X inthe sme way that the paary pe Rodrigue Blact-on-White i a rterion of tha past Pe Moe Rowe would gre that canbe, bu on conn that Phase X has ft seca ped sa deity th rence of he Rodger Bk White atry pes lus ther ol pote and arc yp, ‘MedaghoeTave te fmpesion tha Chang and Rouse ar aking os rater thant eah ste. Rowe ws convinced that the ese systematics of Sramstoisarchaesogy the pling of eulral form inte aed space — At ea be sercd by an inal sty of artifacts 1, to, belie this f be Tove tht th codon obtain, becuse small aoc offer a gear White for ste variation than sch an ray ab 0 setlemen pattern. They prount rete rene of whet Chang has called “relative types Thay bie se meune of ening ard dleeniting beteon the fal, punioular strands of human story. Also, by their very physical sean ad thir oeereeees in lage number oH src ae ame Mibie fo techniques of chronological and atocationl conta tat faphy, geet association, avchiecuabil”asociatios, ie). The ‘Rules on tke eer hand, offers only linited possiblities for “relax tie palo” The constuction of large or special Builings for poliio- Tsou perpos the selection of cemeteries or spell reins or the isp of the dead thse ace ffrts In teaponse to universal human needs cetpaites nd the vesule ave ofen appraiued more readily as “obietve Siresthan ae reatne type “Relatve pes" asc variations do, Pesan, occur in setement par, asin all human ety but they ae Sasly more lic to persivy atlas nthe nial sages of ree Shock ovesigaton any area, The very tat that i the history of a cheetoecalrecarh the eatment of arate has had Tong precedence sean appreciation or even a mention of setlemet peters testimony sry, Chang has become impatient with thik preocapation with ttc sod property so think we cannot have an adequte understand tig af wht wenton inthe past we eschew te socal dimension in pre- tory. But Tauspect that fa hs impatience, Chang as been inlined co pase aver those arceedlogialprocdars That Rowse has spelled out in bis Bets AU last this 1 how T understand thei argument and whee [214/ GoRDON k, wiLLey ‘The remaining papers in the volume ate addressed to various aspects of settlement patterns—as these are seen or inferred archaeologically, or as they are interpreted through ethaograpbie and ethnohistoric data, Deerz: ARTIFACT ARCHAEOLOGY AND CULTURE-CONTACT INFERENCES ‘The paper by Deetz has its central interests in “artifact archacology”— {0 continue with the vocabulary and frame of reference of the Chang Rouse discussions—but these interests are importantly, if indirectly, related to “settlement archaeology.” According to Deets's analysis, a series of op- posite choices is erucial in the maker's selection of attributes during artifact ‘manufacture, By viewing artifacts in this manner, Deetz sees them as in- dicators of nonmaterial culture change, as keys to the processes of culture ‘ontect, a8 the source of data for the reconstruction of technologies, and as more sensitive vehicles of seriation. In his own earlier work (Deets, 1963), he has shown how fine-grained analysis of Arikara pottery can reveal kin ship changes, and how both ceramic analysis and kinship data bear on residence and settlement. In his essay here, his formulations about prebis- toric culture-contact situations also relate to settlement concerns. For ex ample, earlier studies on prehistoric culture contacts (Willey, 1953b; Lathrap, e6., 1956) operated with a settlement concept called “site unit intrusion.” This concept is quite similar to one of the four “attribute con- figurations” that Deetz describes here in his discussion of culture contacts viewed archacologically. But what Deetz has provided is a careful analytic statement about artifacts under the conditions of “site unit intrusion.” There {s a clear implication throughout that “settlement archaeology” and “arti fact archaeology” are equally integral elements of archaeology, {+ ASCHER: DisoRGaNIZATION AND DECAY INTHE SErTLeMENT In his paper, Ascher asks the question: How does a living settlement be- come a part of the past? What are the processes of “disorganization” and ‘decay whereby 2 community of the present is rendered an archacological ‘win? He examines a modem Seri Indian village in an attempt to suggest answers, and notes two such processes. One of these is the progressive “smearing or blending” of refuse and debris. Individual houses and gar- bage dumps recently abandoned show a pattern of concentration, oF dis- Continuous distribution. In contrast, those portions of the village abandoned for a greater length of time reveal a steadily increasing obliteration of the rofuse concentrations, tending toward 2 leveled, continuous distribution of remains. The other process is that of constant re-use and rearrangement of more durable and more favorable building materials or artifacts, These, in being rescued from refuse dumps or automobile geaveyards, are thus carried forward or upward in time, Hea oacluatdegiy Cx pre fetal Bredaed oY nee. ‘Settlement Archaeology: An Appraisal /215 On te face fi these wouk!appens tobe bat ht findings and they wih aby grove De pte treat anton, oy de certain citera and earl environmental conditions But so ft 38 re a among the very fow sytemati aberations 00 He P tel uhechy the pst pte becomes he pst. “Trice: THE DETERMINANTS OF SETTLEMENT PATTERNS thnk, To make tnviiour comparisons forthe various themes taken up ae abviously quite differen; but 1 was particularly pleased o see the theme i of these studies in his introductory statement. One = a ap- proaches. ala environmental or ecological as~ Proof senha th tela he ther fee ten foe bona tendency to view dhe macrosetlement pattern largely in the eco = settlement situation should not be examined a eas ideas “ Crafton nt only ows the sales tthe oust ofthe investigation bat reent pallor is fins \ beaaeli td Commssiy laude aI ] y216/GORDON k. WILLEY sumed under Chang’s microstructure of settlement and the third under his ‘macrostructure. (I take for granted, and T assume that Trigger has aso, that the underpinning of archaeological systematies—modes, lypes, complexes, ‘components, phase formations, and chronological control, all of which Rouse hes insisted upon—has been properly attended to in arriving at levels 2 and 3). Il should be noted that Trigger's three levels of primary observ ion become increasingly more difficult of definition as we proceed from small to large; that is, the individual dwelling or other structure is more readily isolated as a unit for observation than is an individual site. Indeed, theze are many situations (such as the densely settled Viru Valley in Peru) where it becomes something of an arbitrary archaeological decision to de- termine just what is a site unit. In such cases, the archaeologist must pro= ceed as best he can, leaving tentative site-unit definitions open to revision as analysis proceeds. The total landscape distribution, or the macropatiern, is the most difficult of all to comprehend. For one thing, it can be brought into focus only after considerable archaeological research has been carried cout in a zone, region, or area, and after conclusions have been reached about the size and borders of the teritorial unit under consideration, But to return to Trigger's proposals, viewing the settlement data on these three levels of magnitude, he then submits each level to rigorous analysis. For the ievel of the individual house he asks such questions as: What are the conditioning factors of climate and local building resources? How does hhouse construction reflect technological development? In what ways may hhouse size and room arrangement be clues to family structure? What are “. the differentitions in size and quality among houses, and what do. these signify? Is craft or manufacturing specialization indicated in individual houses? Have politcal institutions, such as the nocd for defense, affected house forms? Important to all of this is that the approach be in Taylor (1948) term, “conjunctive”, or, more appropriately, that it takes cogni= zance of what Binford (1962, p. 218) has referced to as the". .. different ‘operational sub-systems of the total cultural system ... ” A consideration ‘of much more than sheer settlement data is needed to provide answers to ‘most of these questions. Trigger offers a good definition for the community, one that can be used fas a measure in appraising archaeological settlement data. To paraphrase him, the community is either the maximal group in face-to-face association Jor any larger unit of stable, more or less continuous settlement representing "TL should emphasize that I do not mean or believe that archaeological sites have “ao realty” or that “they enst only in the mind ofthe clasiir.” T think there isa ood correspondence between site or sellement units and the social unit, the com ‘munity, Ie is a task for arcbaeological research, however, to demonsrate the rela tionship by amplifying. modifying, oF correcting lis conclusions in order to inereare ‘he probabilities of such correspondences, ‘Settlement Archaeology: An Appraisal /217 sabe interaction patterns. At tis level, agin, factors of environment are} | to be considered as determinants, petularly a they bear upon subsistence technology. Family and Linship relations may be rellested Insite pallers, a may wealth and status, The nate of goveramett and religion are ox. pressed in public ullings—tomples, Tors, palaces. Site pateras some- ties also replicate cosmological conceptions "The determinants of Zonal patterns, o macropattern, also include natufale environmental end natural-esouree conditions. What we are contemplating tere isthe phenomenon of the “eulure area,” Trade, particulary in raw 1 materials, helps set limits to a zonal pattern, Political organization, warfare Or the lack of i the invasion of forelgn people, and eigous and ieatonal | factors are all to be considered, oe fe In settlement study on any of these three levels, or in @ multitevel/manner, 4 integration of the data is achieved not only by an examinaiiOW of the con- juntive effets of many determinants but ty the observation of these thvourh time, Ths, 2 community fist etabisked for defense or a «rl lows capital may take on other functions through time—those, fr example, | Seat iaccten cranes of eth nef ca te ommonity Secomes an important and. diversified center of population, | fanctionaly ted formally a ety. In a complementary sense, the data of Settlement are inoprated synchronously by a stud) and demonstration of the relationship between the citi, towns, nd hale of zonal pater. +8 COOK AND HeIZER: FORMULAS FOR POPULATION AND AREA IN THE SETTLEMENT. I “The Cook-Heizer essay isa “formula” attempt to arrive at population fis- es from setlement data, and i promises to he the best such effort in this rection t0 date, The particular approach is foreshadowed only by an ear- lier paper by Naroll (1962), 2 Data from California ethnography are used to | {ic ratio of house Moor space per person, This is computed first for indivi- dual dwvelings, then for total sevlement. The formula was found (0 be stax * ble for aboriginal California; and one cannot help asking if we have here a | tnctioalcortelaon of wie, costal applisbiiy. is question well worth exptoring for other areas and other developmental ranges of A cature. my immediate resetion~no more than e guess~is that it might hold | for hunting-gathering societies and for farming societies up to a cert level of development: boyond this, at perhaps the threshold of urban life, other factors might operate to increase population beyond the ratio of one person’per 20 square feet of floor space. But we do not know, and it may be that the ratio is one that is in some way grounded in basic human biological fof psychological needs and limitations. More investigation along this Tine is tobe awaited with interest + Sea Cok an His (1965 hae , as Att te lorel, i ape Corye nichive i of wn hte CES 4 Wad LT aa mete ARE Lisson boll f tates218/ GORDON R. wiLLEY 2? Waris aNb AYRES: ARCHITECTURAL — CULTURAL CORRELATIONS Whiting and Ayres are also concerned with microstructoral or micro seitement pattern problems and, like Cook and Heizer, their basic datum Fis the individusl éwetting Noor plan, They put the question: If shape and size of floorplan are known, what can be said about other features of cue {ure? This is phrased promisingly for the archacologst, of course, a6 Noor plan data are fequently all that remain to him in ste excavation, Whiting (and Ayres select 136 cultures, on which there are historic or modera data, from the Bthnogeaphic Atlas. Care was taken to select no two caltares that ae in close genetic or dilfusional relationship to each other. Attention was also given t0 seeing that nonagricultural, casual agricultural, and more in. tensive agricultural societies were all included in the sample. Six floorplan forms were reduced 10 two: those essentially curvilinear and those, een, tially rectilineae. The following dwelling structural features were found 9 correlate with the two basic floorplan types: (I) roof forms followed floor forms; @2) pliable materials were used for the walls of curvilinear houses, whereas rigid materials were used for those of rectilinear houses: (3) ret linear houses tended to be larger then curvilinear ones; and (4) rectilinesr houses tended tobe multi-roomed more often than corvilincer ones, Moving feom the strictly architectural to architectaral-caltural core tions, it was observed that large rectilinear, multi-roomed dwellings were correlated with lange extended families, oF ‘with possible high postion in a status hierarchy, or with both. Rectlinear floor plans were also found to be associated more frequently with permanence of setement, wheteat curvilinear plans were correlated with either permanent or nonpermanent settlement To this point, the correlations are what we might expcet. In general, largeness, greater permanence, more demanding construction techniques, ‘etc., relate to rectilinear floor plans, whereas the smaller dvelling, the Himsier structure, the simpler construction process, etc, tend to be assoe- ted with the curvilinear floor plan. Beyond this poin, the causal factors of the observed correlations are more dificult to understand or appraise, For example, the statistical chances are three to one that curvilinear floor plans will be associated with polygynous families and that monogamous [societies are more likely to dvell in rectilinear houses. Whiting and Ayres discus this correlation at some length, and are inclined to accept an esthetic ‘or estheti-Freudian explanation deriving. fcom child-rearing, practices in Polygynous societies,\ My competence 10 evaluate this discussion is cx. tremely limited; [ ean note only that the high correlation between floor. plan curvilinearty and polygyny seems significant, and thet it offers stati tical support for archaeological inferences. The causality of the ‘elation, ship remains, I think, to be explored further, Settlement Archaeology: An Appraisal /219 Scans: Tie SraTE SETTLEMENT ARCHAEOLOGY i : Be cars tackles one ofthe most fascinating and most difficult problems for ao determination of the entity of the state in prehistory. Set- dlement data ste important hee in both the macrostruturl dimension (le | pattem of site distribution over asiable territory) andthe microstructural Imension (he difereniaons beaneca villages, towns, cites, "capitals" | ‘provincial enters" ete) but the approach outlined by Sear i, expiily cr implicily, dependent ona good mary other aspects of archacology. | "In Star paper the state Is defined, with reference to Hocbel, 36 an of anlasd asocation of men for whom 4 specialized suborganzaton funo- thns to transmit state policy into action, of, in somewhat diferent terms, © and with ceference to Gearag, a¢ 2 munber of facetorface comm tava into sovereign poll unit in which there are mechanisms Tor eX siting coetlon over lnese member comannii, over kin group, and over individual members, How ean such phenomena be inferred trom the at- hacaogical past” Sears begins with the definition of the tertory to be omsdefed, ¢ definition based on the presence of highly similar ceramic complexes in aumber of communities within tis territory. The archtco- Topi! Mentification of the "sub-organization” (in Hocbel's terms) of the apparatus of state power is made by demonstrating selllemet recognition Of'a major ceremonial center, city of “epital” within the area of consi tration, which in ils sae, form, and archaeological content ie sigiicatly Giterent from the culturally related and contemporaneous leser centers, towns, villages, or hamlets within the same area, Seas rave example fom the southeastern Unied States an area lacking in good ethnohistoric ot ethnographic rode, end therein lesa weakes ie his argument. is tr pology i cast in an evolutionary fame, At the base isthe “village comma Teel Her, the various sis or communis of he area share the com tmon osramie bond, but the individual settlements are all essentially of the tame paliers- Some may eahibit what appear to be ocremooil srutues, bat none are significantly lager or more elaborate than oer sites of the trea On the second level we see the emergence ofthe “piest sate,” Now, forthe ist time, one site or seitlement of the area bat Decome distinty | lntger and more specialized than the others, Within such a special or cere- tmonial center the suborganization or power sieutue of the sate is identi fied. through specialized regalia found ‘wth important burials. The thind level is the militaristic state" Inthe southeastern United Stats itis 16- | peste by what Sears ha ale the “roplacement conquest sae" 10 ds Trnguish it ftom the more advanced “incorporating conquest stale” a8 seen in the Aztec or Inca Empires, In such a political formation & wartior clas is closely affiiated with leadership, and leadership presents dual aspect of secolaroiitaitie. and religioes functions, Distinguishing. setement Meme siesta220/ GoRDON R, witLEY fetes of the “nltartic sate” ar a single great urban ceremonil een ten adion ster owns and eesnil eter andoticaton. ‘though sympathetic to Sear appoach, I would enter eateat about leg evolulonay schema be the ovetfing consdcation in lasing state “types" do 001 question te geral tend Sear sets ov i roms to Told for the Southeasom United Suter and ao for much of Messmer But at an exception, Tam ot convinced that some ofthe peat Hopeval ceremonial centers sf the Ohio Valley neve not iar cms for pute terior. Seat, however, signs the Hopewelian heizon fo his vilge community” Tere, Dost he Hesiate Hee base fe consider, 2 prior, tet the volitonay level eto low? Or becuse the ttl terry coneemed is rly sal nd among the big Hoperell cenrs there no peri Ula enter that is obvouly paramount over al tbe ees? If so, one sf. inde ofthe Maya Lowlands ofthe Psa and sucundg replons whe ther are great many ceremonial cece of inpresive sce snd magne cence in a flatly mal terry, and where H would be iff to Sige aac end of thing Sears ting to do is exctng but extaoeinar iat, and ik that arhaclgits working Wi this Sort ef macro pater problems wil operate theses of ray low prbsbity or Some tine to come. What is neoed i more obseeatins on eee and otter sapets of clare a thee perain to historeally documented sae formations~the Inca i Perv, the Ae in Mesoamerica, and ata sep dow inthe evftionary sel, the sates or kingdoms of seanth-cenary Cl orb and’ Eevador. Such information would not only promote the sty ot rnp nts nthe pals a es ot wed Suenthen te eorps of general comparative or fentional ology Prehistoric state. ie ze nt deere eeaeee |. VOGT: ZINACANTAN AND THE CLASSIC MAYA ‘The present-day Maya municipio of Znacantan, in the Chiapas High lands, comprises ceremonial coter, oulying pores ot villages, With minor centr, amlts of ual clases of households, and, atthe lowest level of organization, the invidal household isle Vogt ses in i st dement sod tcialsirvture a model forthe old life ofthe Maya in Pre Columbian tines, To be sue, there are diferences. The modern ain cee imoial center or cabecera Is opaized round a Roman Cutholle church {nda tuncpal office representing te Federal Goverament of Mex i the ma religous and governmental center fr the TOD people of the municipio. But tis a "vacant or “ssmi-acan town inhabit largely by thore tial, with their fame, who ar in rlaive residene fer ceremonial of “cargo” poston, It throngs with lif only at times of cere monfes, when tke sostaiing popultions from te outing pars conver Settlement Archaeology: An Appraisal /221 there, Life in the parajes, the water-hole groups, the snas, and the individual id in the an- households is a farming life, and probably goes on much as it cient past, The politico-religious organization binding the muni tether is complex and essentially aboriginal, despite an overlay of Catholic ritual and doctrine, The ceremonies that dramatize this organization appear to have @ symbolic significance, Among these are processionals, which originate within households, are further carried out within hamlets, then move from these to the minor ceremonial centers or shrines of the parajes, and finally to a climax in the main ceremonial center or cabecera. In such processionals, the images of the patron saints of the parajes are carried into the main center; moreover, images from municipios outside of, but tracing ‘history back to, Zinacantan ate brought to the center. Thus, it is easy to sec in these rituals a depiction of social, political, and religious ties lai ut, as it were, on a settlement-pattern map. 'A prominent feature of the Zivacantan system is the “democracy” of its social structure, Perhape a betier way to put it is that there is an absence of aristocracy, and that leadership tends to rotate. Young men hold “cargo” positions for brief periods of time; they then return to their parajes to re- sume farming for several years; later they may be recalled to the ceremon- ial center to serve a term in a more senior position, ete. LF the settlement pattera Vogt observes at Zinacantan is like the settlement pattern found archaeologically in the Maya Lowlands, is it a reasonable ‘working assumption that many elements of sociopolitical and religious Organization seen at Zinacantan can be projected back into the Maya past? One of the most debstable projections or inferences concerns the “demo- cratic” quality of old Maya leadership. There are a namber of tines of ar- cchacological as well as ethnohistoric—evidence that oppose this view (Ruz, 1964); Vogt has discussed some of these. I would also cite the themes of representational art of the Maya Classic Period (A.D. 200-900), and what is known of the hieroglyphic texts of that time, both of which strongly sug- gest aristocratic and even royal lineages (Proskouriakoff, 1960, 1961). How- fever, { have previously argued for the “democratic” view (Willey, 19560), ‘and there is some archaeological support for that interpretation. The ques- tion is a special one for Mayanists, not to be pursued further here. Recent evidence for urban patterns at the Mayan archaeological sites of Tikal and Daibilchaltun will be germane to its solution, as will continued research in 1 good many other aspects of Maya archaeology. The significant thing here is that settlement investigations—both archaeological and ethnographical— have put us in a frame of mind to even ask such a question. Finally, it should be noted that Vogt asks a general theoretical question that none of our other essays have considered: To what extent docs settle- ment pattern influence or determine, rather than merely reflect, social of ganization? Though be does not atterapt to answer the question, there is an222/ GORDON k. WILLEY implication in Vogt’s discussions that settlement may indeed be a causal factor of socal forms. ‘ADAMS: HISTORICAL CORRELATIONS IN A NUBIAN SETTLEMENT Adams’ article is devoted to the study of @ microsettlement pattern in time depth-an examination, through excavation, of the small (I. by 1/2 km) Nubian island of Meinarti located below the Second Cataract of the Nile, He analyzes the natural environmental and eultural constants and de- ferminants of settlement pattern in this setting, Seven levels, or architec: tural phases, are traced from bottom to top, oF in the direction of “time's arrow." The time span involved is twelve centuries, The presentation is ine terpretive as well as descriptive, in that settlement patterning is explained as a register of the more general culcure changes that swept over the whole of Nubie during this span of time. The tightly planned communal buildings and public structures of the first seitlement are seen as « reflection of the stable Meroitic polity; the helter-skelter assemblages and the evidences of temple destruction of the second level are taken as a measure of troubled times and a breakdown of central authority; subsequent planned commu nities and alternating periods of disorganization are interpreted similarly; |, anda final defensive architectural phase bespeaks war and invasion. ‘One comment should be made about the interpretations. They are made in a well-controlled, well-documented framework of Nubian’ prehistory ‘and history. Thus, when Adams pairs tight architectural planning on the litle island of Meinarti with centralized political authority, he is making 4 historical correlation, Is it also a functional one? And, conversely, is inregularty of settlement and architecture and lack of centralized authority ‘another functional correlation, or merely a historical one? { am inclined t0 believe there are functional correlates or causal interrelationships. here But this is one of the things we cannot be sure about yet in seitlement inter. pretation, ‘Aspects oF SETTLEMENT PATTERNS-CONCLUSIONS IN RETROSPECT In retrospect, and briefly, what are the main things these essays have to tellus? ‘The Chang-Rouse controversy is a kind of archaeological “jurisdiction- al” dispute. That these two able archacologists felt strongly about it ine dicates, perhaps, that such a dispute needed aicing. 1 consider it a “given” that certain archacological procedures should precede settlement pattern study; and I would have thought Chang also assumed this, but Rouse has interpreted him otherwise. I believe Chang sought primarily to inject an awareness of settlement pattern into archaeological investigation at as ‘early a point inthe procedures as feasible. | j i i i | | i i | j ; 4 i j Settlement Archaeology: An Appraisal /223 Deecz has made more sophisticated our ways of thinking about atact and their determinants This isa primary step to the ute of artifact analysis fa means to stawer questions aboet nonmaterial cltre, Such analysis has been, and maybe coordinated with setlement 0 "Ascher asks archaeologists 0 consider how living commanites eed the realm of prehistory: What are the meshenical and chemical ways this transformation took place, and through what culttal and natural omnis faculty at Berkeley from 1928 until his retirement in 1964. He has published widely in physiology and anthropology, and is frequent col laborator with Dr. Robert F. Heizer in archaeological research and writings. 27