historical inconsistencies: huron longhouse length, hearth ... · historical inconsistencies: huron...

HISTORICAL INCONSISTENCIES:

HURON LONGHOUSE LENGTH, HEARTH

NUMBER, AND TIME

Colin Varley and Aubrey Cannon

Data from three Middle Ontario Iroquoian periodvillages in Simcoe County reveal an anomalouspattern of hearth spacing. They show greatervariability and longer average spacing than isnormally assumed for Ontario Iroquoian longhouses.We interpret this as an exaggerated expression ofthe symbolism of power and prestige normallyassociated with large lineages and longer houses.

INTRODUCTION

One of the standard assumptions of Iroquoianarchaeology is that there is a more or less constantrelationship between house length and hearthnumber. It is generally recognised that as longhousesize increases, so do the number of hearths locatedalong the central corridor of the house. As anempirical generalization this relationship is wellestablished, and Dodd (1984:256-257) has demon-strated its validity from the Early Ontario Iroquoisperiod through to the Historic period Huron. Dodd's(1984:274) analysis showed that average hearthspacing, while low in the early period (1.9 m), wasrelatively consistent after A.D. 1450 (3.6 m in A.D.1450-1609, and 2.9 m in A.D. 1609-1650). Only in theMiddle Iroquoian period was there a significantdeviation from this pattern, when average hearthspacing increased to 5.3 m.

Unfortunately, there has been a tendency to usethese observed regularities in efforts to makeempirical data conform to standard expectations.When hearths are not actually observed they aresometimes "inferred" on the basis of an assumptionthat hearth spacing was generally consistentthroughout Iroquoian prehistory (e.g., Finlayson1985:107). The problems inherent in this practice arecompounded when hearth numbers are used to inferpopulation levels based on the ethnographic observa-tion of two families per hearth and an assump-

tion of uniform family size (e.g., Finlayson

1985:107,109).In criticizing the assumption of a

constant relationship between house lengthand hearth number Warrick (1990:89-90,225-228) outlines several factors which caninfluence the number and location ofhearths within a given residence area. Theseinclude variable hearth preservation(particularly poor in sites that have beensubject to ploughing), "hearth drift" (singlehearths changing locations over their periodof use), and contemporaneity of hearths(multiple hearths in use at different times).The problem with predicting hearth numberfrom house length, however, goes beyondthe potential for error in making hearthcounts or infer-ring population size. Theimposition of standard assumptions on thearchaeological record also precludes therecognition of non-conformities which maybe among the more informative aspects of

the archaeological record. By way ofillustration, we document deviations fromthe standard relationship between houselength and hearth number which appear torepresent social distinctions withinindividual villages and suggest a period ofsignificant social change during the lateMiddleport and Lalonde periods in SimcoeCounty (circa A.D. 1380-1520).

DATA

Houses from the Carson, Baumann, andWiacek sites in Simcoe County (Figure 1)illustrate a distinctive pattern of hearthspacing within longhouses. The earliestvillage of the three is the Wiacek site, a lateMiddleport village south of Barrie (Lennox etal. 1986). Portions of five houses wereexcavated, but only two houses werecompletely uncovered (Figure 2). House Onemeasures 41 m in length and shows evidenceof seven hearths, al-though two aresufficiently close to be consid-

VARLEY AND CANNON HURON LONGHOUSE LENGTH, HEARTH NUMBER, AND TIME 85

Figure 1. Location of Wiacek, Carson, and Baumann Sites.

86 ONTARIO ARCHAEOLOGY No. 58, 1994

ered a single hearth cluster. Average hearthspacing ranges from seven metres per hearthfor six distinct hearths to six metres perhearth for seven distinct hearths. House Twomeasures 36 m in length and contains fourhearths. The average hearth spacing for thishouse is seven metres per hearth. Theevidence suggests that the Wiacek site housestend to have greater hearth spacing for aMiddle Iroquoian village than Dodds(1984:274) average of 5.3 m. While this doesnot seem a significant deviation, in thecontext of later houses in the region, it pointsto an emerging period of disconformity in thestandard pattern documented for Iroquoianhouses.

The Carson site is a late fifteenth or earlysixteenth century (C¹4 A.D. 1507±27) unpal-isaded Lalonde village located just outsideBarrie, Ontario (Figure 3). Excavations con-ducted by Archaeological Research Associatesin the summer of 1989 uncovered the remainsof eight longhouses, ranging in length from 52to 75 m. Of these eight, six were completelyexcavated. House Four was severelydamaged by previous gravelling operations,and the season ended before House Eightcould be completely excavated. Housefeatures have excellent preservation, andhearths are clearly evidenced by large areasof fire reddened soil and numerous small ashfilled post-molds within the reddened areas.Recorded hearth depths range from 4 to 22cm below the subsoil surface. We areconfident that the recorded hearth numbersare accurate (Figure 4). De-spite the qualityof preservation, however, only three to fiveindividual hearths were recorded for eachhouse (Table 1). The minimum aver-agehearth spacing is evident in House One at onehearth every 8.7 m. The greatest averagehearth spacing is found in House Two, at 13m. The Carson houses clearly exhibit greaterhearth spacing than is documented for othertime periods, or for the Wiacek site.

Viewed in isolation the Carson site couldmerely be an anomalous village, but furtherevidence of a discernable historical trend ispresent in the single house excavated atBaumann, a fifteenth century Lalonde sitenear Warminster. House One at Baumannmeasures roughly 66 m in length (Stopp1985:6). Four large irregular areas ofreddened soil were identified as hearth areaswithin this house (Stopp 1985:8). Based onthese figures the average hearth spacing atthe Baumann site is 13 m. Although a singlehouse cannot be

Table 1. Carson Site Houses.

House # Length(m)

# ofHearths

AverageInterval (m)

1 52 5 8.7

2 52 3 13.0

3 59 5 9.8

5 52 4 10.4

6 57 4 11.4

7 57 4 11.4

considered representative of the site as a whole,the similarity between this house and those fromthe Carson site suggest a significant deviation fromthe norms established for other time periods.

It is important to make clear that we are notsuggesting that hearths are located at regularintervals within a house (Figure 5). Rather, ourestimates of hearth spacing are based on anaverage amount of available space per hearth. Weare also not suggesting that each hearth was usedas frequently (or for as long) as the rest of thehearths within a house. A variable history of hearthuse within individual houses has been documentedrecently for the late sixteenth century Benson siteHuron village (Fogt and Ramsden 1993). Evidencefrom the St. Lawrence Iroquoian Lanoraie site alsosuggests that some hearths may have been centresfor different activities, and that one hearth mayhave played a central role in the household(Clermont et al. 1983:133). The irregular spacing ofhearths within a house, however, allows for thepotential of future hearth placement within anestablished frame-work of house arrangements.

Our data suggest that not only are Lalonde erahouses very large, they are also much less denselypopulated than longhouses in other periods.During the periods before and after Lalonde, thegeneral trend of house length and hearth numberco-variance essentially con-forms to the normalpattern in effect through-out most of prehistory,although as we have already noted, MiddleIroquoian villages generally tend to have greaterhearth spacing. Our data demonstrate, however,that in the



Lalonde period there is a very distinctdisconformity in the relationship between hearthnumber and house length that requiresexplanation.

DISCUSSION

A number of suggestions have been put forwardto account for the dramatic growth in the length ofIroquoian longhouses. One suggestion is thatrestricted corporate group control over economicresources and trade favoured larger group size(Hayden and Can-non 1982:149). Specificreference has been made to the enhancedimportance of trade in the period of Europeancontact (Trigger 1976:421-23). However, corporategroup size and membership do not directlyaccount for the effort to maintain commonresidence, beyond perhaps as an expression ofgroup solidarity. Control over trade may also beless important in the case of Huron longhousessince houses actually become shorter during theperiod when competition for European trade wasunderway (Warrick 1984:75).

A more mundane explanation for the generallengthening of longhouses is that it was primarilya means to conserve construction materials andheating fuel, while managing to house greaternumbers of people (Wallace 1971:86, cited in Dodd1984:216). In light of this explanation, however, wewould expect the relationship between hearthnumber (and by implication population) and houselength to remain constant. To the contrary, ourdata indicate that house length was actually ex-ceeding the growth in numbers of people beinghoused. If heating and material efficiency wereconcerns, then the pattern we observe for theMiddle Iroquoian period in general and for theCarson and Baumann sites in particularrepresents a waste of space and resources.Subsequent reduction of longhouse length and acorresponding increase in the number of houses inthe Protohistoric and Historic periods is alsodifficult to explain if conservation of materials wasan overriding concern.

The pattern of longhouse development may havebeen much more closely related to the role of thelonghouse as a symbolic material expression ofsocial structure and dynamics. An important socialand political unit of Huron society was the lineagegroup which was

known historically to reside together in onehouse. There is also evidence to suggest thatgreater social and political prestige was givento lineages with the most members (Trigger1976:57). Large lineages were better able toassert their political and social will at theexpense of smaller lineages. In cases of mur-der, for example, in which the accused per-son's relatives were expected to contributetoward payment of the fine (Thwaites 1896-1901:10:215-217), a large lineage wouldensure adequate resources for compensation.At the same time, the fine paid might only bemoderate if the murdered person's relativeswere considered obscure and of little account(Thwaites 1896-1901:13:15). Since lineagemembers lived together, larger lineages wouldrequire longer houses. Trigger (1990:126)suggests that the development of very largehouseholds may reflect lineage rivalry withincommunities formed by the amalgamation ofsmaller ones. We suggest that the relationshipbetween house length and family size alsoresulted in house size becoming as much asymbolic expression of social and politicalstatus or power as it was a reflection oflineage size.

The fifteenth century deviation from thestandard ratio of house length to hearthnumber can be viewed as the use of inflatedlong-house length to assert socio-politicalaspirations in excess of a reality warrantedby the actual number of lineage members. Inessence, we see the Lalonde period develop-ments as representing a symbolic competitionfor status and status recognition amonginfra-village lineages. Houses becamesymbolic material expressions that werephysically empty, as individual housepopulations were more thinly distributedwithin the larger houses.

There are a number of specific reasons whylonghouse length may have become exagger-ated beyond the spatial requirements of itsmembers. One possibility is the deliberate orsubconscious manipulation of the symbolismof house length to create a false impression oflineage size and status, which might be suffi-cient to enhance prestige and influence. Inthis sense, longhouses would be equivalentto any ostentatious material display, whichthrough competition becomes exaggeratedbeyond its basis in social reality. However,house size alone is a weak substitute forlineage status


and power, which is more directly related to

the actual size of lineage membership.Alternatively, longer houses may have been

built with the anticipation or intention of actu-ally increasing the size of household member-ship. As Fogt and Ramsden (1993) have shownfor the Benson site, the developmental historyof a longhouse may typically involve changesin the number of residents, with houses builtto accommodate such changes. The muchlonger but much more empty houses of theLalonde period might simply representhonest, though grossly overoptimisticprojections of future household membership.Clearly, if all lineage groups are competing formembership and making house spaceavailable for anticipated future members, theinevitable result would be a greater averagevolume of empty space. It is also possible thatthe impression of lineage size and statusconveyed by exaggerated house length wasintended as an inducement for would-bemembers to join. Since new villagers would bemore attracted to established lineages thatappeared to be more prestigious, and ifcompetition for new lineage members washigh, it would be a good strategy to enhancethe impression of a large lineage and at thesame time have space readily available toaccommodate any new members.

A variety of specific individual or groupmotives can generate a common historicaltrend. In this case, we have suggesteddeliberate or subconscious manipulation oflong-house symbolism to produce anessentially false impression, an honest butoverly optimistic projection of future growthin household membership, or a symbolicenhancement of household prestige and realspatial accommodation designed to attractnew members. Other possibilitiesundoubtedly exist, but an essential feature ofall of these suggestions is the sociallycompetitive basis for increasing longhouselength to a point well beyond the spatialrequirements of the actual number ofresidents. The inevitable result of these devel-opments was that when houses could no lon-ger increase in size, and when the disparitybetween ideal and reality was readily appar-ent, they eventually returned to dimensionsmore suited to residents' needs.

By the Historic period, villages are not justlarger in area, but contain significantly more(and shorter) houses than in earlier periods(Figure 6, cf. Warrick 1984:60-61). In thisperiod

there is a return to the normal relationshipbetween house length and hearth number, which,if the considerations outlined above hold, suggeststhat longer houses were no longer effective assymbols of social or political prestige.

Steckley (1987:21) has shown that, among theHistoric period Huron, suggesting that someonelived in "an empty house" was an insult. We seethis as the historical outcome of earlier, prehistorictrends toward exaggerated house length andaspirations toward unfounded lineagemembership and power. The insult "to live in anempty house" would, in effect, be mockingunfounded social and political aspirations appliedto household groups that either could notmaintain their membership in competition withother groups or were seeking to create anunwarranted material impression. The effect ofthis insult, and the social conditions whichbrought it about, may have been to re-establish thenorm of house length to hearth number ratio,which was increasingly important given thephysical constraints of increases in village sizeand house numbers. In this context, it would not besurprising if the material expression of lineageprestige without foundation had become thesubject of public ridicule.

CONCLUSIONS

Iroquoian house length and its changes overtime may have been strongly influenced by theessential symbolism of the longhouse, al-though italso may have been related to the number ofresidents and, by implication, the status andsocial power of the lineage group. In our view, theMiddle Iroquoian period represents an elaborationof the symbolic dimension alone, since accordingto hearth counts there were the same number ofpeople in a much larger space. Extending into theLalonde period, the Carson and Baumann sitesseem to represent the apogee of symbolicexpression through the medium of longhouselength. At this point the exaggerated extremes ofhouse length and the redundancy of expressionbetween lineage groups had diminished thesymbolic value of longhouse length. The result wasa rapid return to the normal ratio of hearth numberto house length and a dramatic de-crease inoverall length into the historic period.

Moreover, we suggest that the pattern ofincrease and decrease in Iroquoian house length istypical of a general pattern that occurs whenmaterial culture is used as an expression of status(Cannon 1989). A material medium that is used toexpress distinctions of status will typically gothrough a period of elaboration in response tocompetitive pressures, represented in this case bythe lengthening of houses. This culminates in apeak of ostentation, which for longhouses is thepoint at which house length goes well beyond itsnormal relationship to the number of occupants.The result of reaching this point of expressiveredundancy (Cannon 1989:437-438) is that themedium loses its expressive value and rapidly goesout of use as a symbolic expression, represented inthis case by the decrease in house size. The finalstage typically involves the use of public ridicule tokeep laggards from socially inappropriate andanachronistic expressions of would-be status.Similar instances of symbolic elaboration andcollapse can be found in examples ranging fromEnglish gravestones (Cannon 1989) to Neolithicpottery decoration (Hodder 1990).

Beyond furthering our understanding of Huronprehistory we hope that this paper also encouragesothers to re-examine the particular aspects of thearchaeological record which are sometimesneglected when the greater concern is withestablishing general patterns and trends. To movebeyond simply reinforcing what we think wealready know about Huron prehistory we must bewilling to examine deviations in the archaeologicalrecord as important subjects in themselves. Non-conformities are not simply accidents or historicalanomalies; they are valid and important historicaldevelopments. If we are to understand the past wemust be willing to accept and wrestle with thevariable realities we observe instead of imposingthe normative generalizations we construct.

REFERENCES CITED

Cannon, A.1989 The Historical Dimension in Mortuary

Expressions of Status and Sentiment.Current Anthropology 30:437-458.

Clermont, N., C. Chapdelaine, and G. Bane1983 Le Site Iroquoien de Lanoraie: Tern-

oignage dune Maison-Longue. Rech-erches Amerindiennes au Quebec,Montreal.

Dodd, C.1984 Ontario Iroquois Tradition Long-

houses. National Museum of Man,Mercury Series, Paper 124:181-437.Archaeological Survey of Canada,Ottawa.

Finlayson, W.1985 The 1975 and 1978 Rescue Excava-

tions at the Draper Site: Introductionand Settlement Patterns. NationalMuseum of Man, Mercury Series,Paper 130. Archaeological Survey ofCanada, Ottawa.

Fogt, L., and P. Ramsden1993 From Timepiece to Time Machine:

Scale and Complexity in IroquoianArchaeology. In Debating Complex-ity, edited by D. Meyer and P.Dawson. Proceedings of the 26thAnnual Chacmool Conference.University of Calgary, Calgary. inpress.

Hayden B., and A. Cannon

1982 The Corporate Group as an Archaeo-

logical Unit. Journal of Anthropologi-

cal Archaeology 1:132-158.Hodder, I

1990 The Domestication of Europe: Struc-ture and Contingency in NeolithicStudies. Basil Blackwell, London.

Lennox, P., C. Dodd, and C. Murphy1986 The Wiacek Site. Ministry of

Transportation and Communications,London, Ontario.

Steckley, J.

1987 An Ethnolinguistic Look at the Huron

Longhouse. Ontario Archaeology

47:19-32.Stopp, M.

1982 An Archaeological Examination of theBaumann Site, A 15th CenturySettlement in Simcoe County,Ontario. Ontario Archaeology 43:3-29.

Thwaites, R. G. (editor)1896-1901 The Jesuit Relations and Allied

Documents 73 Volumes. BurrowsBrothers, Cleveland.


Trigger, B. G.1976 The Children of Aataentsic: A History

of the Huron People to 1660. McGill-Queen's University Press, Montrealand Kingston.

1990 Maintaining Economic Equality inOpposition to Complexity: An Iro-quoian Case Study. In The Evolution

of Political Systems: Sociopolitics inSmall-Scale Sedentary Societies,edited by S. Upham, pp. 119-145,Cambridge University Press, Cam-bridge.

Varley, C.1993 The Carson Site. In North and South: Two

Views of the Black Creek-Lalonde Period,edited by P. Ramsden, pp. 64-107.Copetown Press, Dundas, Ontario.

Wallace, B. J.1971 Village Life in Insular Southeast Asia.

Little, Brown, and Company, Boston.Warrick, G.

1984 Reconstructing Ontario Iroquoian VillageOrganization. National Museum of Man,Mercury Series, Paper 124:1-180.Archaeological Survey of Canada, Ottawa.

1990 A Population History of the Huron-Petun,

A.D. 900-1650. Unpublished Ph.D.dissertation, Department of Anthropology,McGill University, Montreal.

Colin Varley and Aubrey Cannon McMasterUniversity, Hamilton, Ontario.
