PUBLICATIONS :: NYSM RECORD :: Current Research in New York Archaeology

cover Current Research in New York Archaeology: A.D. 700–1300

edited by Christina B. Rieth and John P. Hart

Chapter 2: The Carpenter Brook Site, Social Setting, and Early Late Woodland Ceramic Vessel Variation in Central New York

Donald A. Smith

Abstract. The chapter hypothesizes that the social setting(s) in which early Late Woodland potters in central New York intended vessels to be used is reflected in some of their pots’ attributes. Characteristics of the ceramic assemblage from the Carpenter Brook site, which was probably deposited during ritual, are compared with those of pottery from three village sites: Bates, Maxon-Derby, and Sackett. On average, relative to the vessels from the villages, the Carpenter Brook pots are larger, have walls with characteristics that would have made them less durable as cooking vessels, and have larger and more complex decoration. The differences are possibly related to the different functions of pottery in ritual contexts, relative to how it was used in more prosaic social settings. The possibility that the site was related to ceramic vessel production is also discussed.

This chapter explores the possibility that the social setting(s) in which late Middle Woodland–early Late Woodland (ca. A.D. 900–1300) potters in central New York intended their vessels to be used played a role in their decisions concerning the mechanical and decorative attributes of their wares. I focus on the assemblage from the Carpenter Brook site in Onondaga County, excavated by William Ritchie in 1946 (Figure 2.1) (Ritchie 1946). Ritchie argued that the site, which played an important role in his formulation of the “Owasco” / early Late Woodland (A.D. 1000–1300) culture-historic sequence, was deposited during a series of ceremonial events (Ritchie 1947; Ritchie and MacNeish 1949). This paper compares several attributes of the pots from Carpenter Brook, including their diameters, wall thicknesses, and the size and complexity of their decoration with those of vessels from three early Late Woodland central New York village sites—Bates, Maxon-Derby, and Sackett—which were likely deposited in more prosaic contexts. The results indicate there are differences between the pots from Carpenter Brook and those from the villages that may reflect the differing needs of the distinct social settings (i.e. formal ritual vs. “everyday” or prosaic) in which they were used. Specifically, the distribution of vessel attributes from Carpenter Brook relative to those from the villages is consistent with the hypothesis that, in general, the potters who made the Carpenter Brook pots expended greater amounts of effort on their vessels’ decoration and less effort on characteristics that would have ensured the pots had long use-lives as cooking vessels. However, other interpretations are certainly conceivable, and the possibility that Carpenter Brook was associated with ceramic production or that some of the differences between its pots and those from the villages were related to changes through time are also addressed.

Figure 2.1
Locations of Carpenter Brook and the village sites used in this study.


The Carpenter Brook site was located in the bank of the meandering stream with which it shares its name, several hundred feet south of its confluence with the Seneca River. It was initially discovered by avocational archaeologist Earl Mann in 1922, who subsequently informed Ritchie of the find (Mann 1922–1946; Ritchie 1947:56). It comprised a single artifact deposit measuring about 12 m by 3 m (40 ft by 8 ft) that was eroding from the east side of the stream bank roughly 60 to 100 cm below the surface of the adjacent terrain (Figure 2.2) (Ritchie 1947:56, 58). The stream bank that contained the deposit was near the base of a sandy knoll that rose just east of the brook. Ritchie (1947:56) describes three soils overlying the artifacts. The uppermost was a 46 to 53-cm-thick stratum of “fine brown culturally sterile silt ... [that] washed down from the closely adjacent knoll.” The silt overlaid a layer made up of lenses of “brown silt, coarse gray or buff-colored sand, and fine calcareous sinter” that totaled 13 to 25 cm in thickness. Ritchie attributes the varying qualities of this layer to changes in depositional mechanisms: “It evidently represented wash from the knoll and stream[-]deposited sand and tufa or calcium carbonate, laid down in periods of high water.” As with the overlying silt, he found no artifacts in it. Finally, below this was a 20 to 30-cm-thick “stratum of coarse sand and fine gravel,” the lower 15 cm of which contained the artifact deposit. The sand/gravel layer was “so heavily interspersed with masses of tufa as to appear a veritable zone of nodular calcareous sinter, stained and streaked with limonite.” Ritchie notes it likely accumulated in a submerged depositional environment. He describes the artifact deposit as:

Figure 2.2
The Carpenter Brook deposit during Ritchie’s excavation (Ritchie 1947:61).

a nearly solid mass of potsherds, lying among scattered water-worn boulders similar to those now littering the stream bed, and like them encrusted with calcareous sinter or tufa deposited from solution in the cold spring-fed water of the brook. Present among the sherds were occasional animal bones, a few mussel and other shells, and very rarely an artifact of other type, all coated in sinter varying in thickness from a mere white film of 2–3 mm to a heavy encrustation up to 2 cm. (Ritchie 1947:56–58)

The artifacts were underlaid by “the old stream bed, composed of coarse light gray or buff-colored sand and fine gravel” (Ritchie 1947:58). Ritchie (1947:58, 66) attributed the exposure of the site to erosion resulting from a decrease in the water level of Carpenter Brook caused by nineteenth-century attempts by the state to drain the nearby Montezuma Marshes.

In addition to the soils described by Ritchie, Mann (1922–1946) mentions that he observed “a layer of very fine clay” in the “creek bed.” While Ritchie does not mention such a deposit, the USDA soil survey for Onondaga County indicates that two of the soil types found along Carpenter Brook (Teel silt loam and Williamson silt loam) occasionally contain lenses of clay, suggesting Mann’s observation may have been accurate (Hutton and Rice 1977:100, 110; Soil Survey Staff 2011).

During his excavations, Ritchie (1947:58) tested the surrounding area with a series of test pits, and a trench measuring 1.2 m (4 ft) wide and at least 2.4 m (8 ft) long that extended east from the artifact deposit toward the base of the nearby knoll. None of this testing yielded any additional prehistoric cultural material. Mann, who visited the site occasionally until 1946, also searched nearby for signs of prehistoric occupation or use. He identified “only slight signs of an earlier occupation on the top of the adjacent knoll,” but unfortunately he makes no further observations as to the characteristics of these remains (Mann 1922–1946:4).

The artifact assemblage from the site, now in the collection at the Rochester Museum and Science Center, is dominated by pottery; Ritchie recovered hundreds of vessel fragments including 437 rim pieces from at least 125 pots (Figure 2.3) (Ritchie 1947:64; Ritchie and MacNeish 1949:118). He believed the sherds were from whole vessels that were broken in situ; Mann’s notes include a similar observation (Mann 1922–1946; Ritchie 1947:64–66). In addition to the sherds recovered by Ritchie, Mann collected pieces from up to another 25 vessels during his visits and Ritchie estimates a further 50 pots were lost due to erosion of the stream bank, bringing the total number of pots from the site to about 200. (Ritchie 1947:64).

Figure 2.3

Reconstructed rims of Carpenter Brook vessels.

Mann’s notes also include some comments concerning characteristics of the pots he collected. He remarks that vessels of all sizes were present, ranging from “a tiny bowl … to immense bowls nearly an inch thick.” He observes that the amount of decoration on the pots was similarly variable; while some vessels had minimal decoration, others had “impressed designs of some intricacy, well down the sides” (Mann 1922–1946:2–3). Both Ritchie and Mann remarked that the sherds in the deposit were highly fragile, and many disintegrated when they attempted to collect them. Ritchie (1947:60) attributed their fragility to submersion in water: “an unusual feature is their friability, doubtless to be accounted for by long immersion in cold water.” He goes on to note that Sagard’s seventeenth-century description of Huron pottery included the observation that “they [the pots] cannot stand moisture and cold water for long, but become soft and break at the least blow given them” (Ritchie 1947:60; Sagard 1939:109). Mann (1922–1946) provides additional hypotheses for the fragility of the sherds: they were “either frail due to poor baking or the temper being drawn by the lime and moisture.” He also states that “in some cases [they do] not appear to have been used.”

In addition to the pottery, Ritchie found a total of 150 animal bones and bone fragments scattered among the ceramic fragments. Of these, 90 (60 percent) were from a minimum of 7 Black bears (Ursus americanus) and 19 (a further 13 percent) were from deer (Figure 2.4). Almost all the ursine bones (79 out of the total 90) were from either the bears’ heads or feet. The remainder of the faunal assemblage included material from woodchuck, rabbit, raccoon, turkey, muskrat, dog, mink, beaver, puma, and one bone from a fish. Mann also noted the presence of bear and deer remains as well as those from other animals, he does not enumerate (Mann 1922–1946; Ritchie 1947:62).

Figure 2.4

Bear mandible from Carpenter Brook.

Besides the pottery and faunal material, Ritchie found just 20 additional artifacts. Among these was a 5.7-cm-long by 3.5-cm-wide fragment of a “unique effigy of the human phallus” made from potters’ clay and detailed with a cord-wrapped stick (the same type of implement typically used to decorate early Late Woodland pottery) (Figure 2.5) (Ritchie 1947:63). The phallus also includes an anatomically accurate longitudinal perforation; as Ritchie describes:

Unusual interest attaches to the fact that the true position of the urethra ... is correctly indicated by a small perforation, 1.5 mm in diameter, which can be followed by the probe almost to the extremity of the glans. However, this portion of the model was structurally too weak to carry the desired larger perforation, so the urethral passage, 5 mm in diameter, was placed above it through the center of the object. The existence of this sole anatomical flaw creates the impression that the urethra was intended to serve some functional purpose. Speculation on this point is largely precluded by the absence of the proximal portion and the lack of comparable specimens. (Ritchie 1947:63–64)

Figure 2.5

Clay phallic effigy (probably smoking pipe) from Carpenter Brook.

Although Ritchie does not comment on the function of this object, Parker (1922:197) reports on a similar item—a “phallus in clay”—from the Late Prehistoric Richmond Mills site in Ontario County that he interpreted as a smoking pipe. Engelbrecht (personal communication 2004) suggests the Carpenter Brook phallus was probably used for the same purpose.

The remaining artifacts Ritchie collected from the site include: a complete obtuse-angle clay smoking pipe; an Onondaga chert core; two sandstone pebbles; 3 sandstone hammerstones; 10 Onondaga chert flakes; an Onondaga chert pebble; and a possibly-polished white quartzite pebble. In his notes, Ritchie (1946) mentions that he observed “large charcoal granules and fragments,” as well as some fire-cracked rocked, but he did not retain any of this material. His notes also indicate that he collected a piece of unburned wood, but it is no longer present in the Rochester collection. Mann (1922–1946:1) reports on two additional items he collected: an antler ‘prong’ and a “very crude adz-like implement of sandstone” he suggests may have been related to pottery production. Although he states he gave the adz-like item to Ritchie, it is not present in the Rochester Carpenter Brook collection.


Results from recent research concerning Middle and early Late Woodland (A.D. 1–1300) pottery chronology, settlement, and subsistence in the Northeast (e.g. Gates St-Pierre 2001; Hart 1999, 2000; Hart and Brumbach 2003, 2005, 2009; Hart et al. 2003; Prezzano 1988; Schulenberg 2002a) necessitate a brief discussion of the changing interpretations of Carpenter Brook’s age. To accommodate methodological developments and new data Ritchie’s ideas about the age of the site changed several times in the years after he excavated there (Figure 2.6). All his estimates were based on qualities of the site’s ceramic assemblage. Initially, he believed the site belonged in the Canandaigua Focus, a taxonomic entity that he speculated in his dissertation lasted from about A.D. 1200 to A.D. 1450 (Ritchie 1944:13, 28–30; 1947:64, 67). Later, he and MacNeish reconsidered the age of Carpenter Brook relative to Canandaigua, and argued that Carpenter Brook was older (Ritchie and MacNeish 1949:118). Ultimately, Ritchie believed the site was contemporaneous with the Maxon-Derby village site in the village of Jordan, for which he acquired uncalibrated radiocarbon dates of 850±100 B.P. and 850±150 B.P. (A.D. 1100±100 and A.D. 1100±150), respectively (Ritchie 1980:275; Ritchie and Funk 1973:210). He based his argument for the contemporeneity of the two sites on their proximity (they were separated only by about two km) and similarities in their ceramic assemblages. In fact, he speculated that people from Maxon-Derby “were probably intimately connected” with those who visited Carpenter Brook (Ritchie and Funk 1973:195, 210).

Figure 2.6

Changing interpretations of the age of Carpenter Brook.

However, recent AMS (accelerator mass spectrometer) and radiometric dates on charcoal from Maxon-Derby obtained by Hart (2000:8–13, 17) from samples collected by Ritchie indicate that in addition to a calibrated eleventh-century occupation corresponding to Ritchie’s uncalibrated A.D. 1100 date, the site also had an occupation from the cal. mid-twelfth to mid-thirteenth centuries A.D. Beyond this, Schulenberg (2002a) and Hart and Brumbach (2003:743–745; 2005) have acquired AMS dates on charred cooking residue adhering to sherds of Middle and early Late Woodland pottery types that are far outside the time ranges Ritchie assigned them (see also Gates St-Pierre 2001). Beyond demonstrating that Ritchie’s “Owasco” ceramic types lack the temporal sensitivity he believed they embodied, these studies have substantially revised many long-held ideas concerning regional cultural developments during the Middle and Late Woodland in New York. For example, Hart and Brumbach (2003) have shown that none of the traits that Ritchie argued appeared together around A.D. 1000, including nucleated villages, longhouses, and a system of agriculture based on maize, beans, and squash, appeared at that time. Instead, squash and maize have been found in much earlier contexts, “while beans, maize-beans-squash agriculture, longhouses and associated matrilocality, and villages are later” (Hart and Brumbach 2003:746). This research also necessitates that Ritchie's estimation of Carpenter Brook’s age be reconsidered.

To address this issue, I obtained AMS dates for charred residue adhered to two sherds from Carpenter Brook, one of which has the qualities of Carpenter Brook Cord-on-Cord in the Ritchie and MacNeish (1949) typology and the other is Owasco Corded Horizontal (only about five of the sherds from the site have adhered residue). The dates are: 1010±40 B.P. (cal. 2σ range and median probability of A.D. 970 [1030] 1060 [p = .71] and A.D. 1080 to 1160 [p = .26]) (Beta-193706) for the Carpenter Brook Cord-on-Cord sherd and 1100±40 B.P. (cal. 2σ range and median probability of A.D. 870 [920, 980] 1020 [p = .99]) (Beta-193707) for the Owasco Corded Horizontal piece. The dates are not significantly different at the 95 percent level of certainty and have a pooled mean of 1055±28 B.P. (calibrated 2σ range and median probability of A.D. 900 to 920 [p = .12] and A.D. 940 [1000] 1020 [p = .88]). (Calibrations and calculations were completed with Calib v. 6.0.1 [Reimer et al. 2009; Stuiver and Reimer 1993, 2010].) These dates indicate Carpenter Brook is roughly 100 years older than Ritchie believed. Unlike many of the cases reported by Schulenberg and Hart and Brumbach where sites were used more than once, the Carpenter Brook dates do not indicate it was visited during a time outside its primary period of deposition (although more dates for the site may change this).

Carpenter Brook, its Pottery, and Ritual

Ritchie’s interpretation of Carpenter Brook was partly influenced by its atypical qualities relative to those of other large prehistoric artifact deposits. “We do not,” he notes, “abandon a secure position in formulating the initial premise ... that the pottery dump is no normal refuse midden” (Ritchie 1947:67). This observation is based on two of Carpenter Brook’s qualities: its distance from other sites and the unusual composition of its artifact assemblage. First, with the exception of the “slight” indications of earlier prehistoric use Mann (1922–1946) identified on the adjacent knoll, neither he nor Ritchie located any nearby evidence of additional prehistoric use. In fact, in his Carpenter Brook monograph, Ritchie (1947:67) notes that the site was at least 500 meters from the nearest known habitation site, that at the Felix farm on the bank of the Seneca River (at the time of his monograph, Ritchie believed Carpenter Brook was deposited by people living at Felix). He implies it is unlikely that the inhabitants of Felix would have traveled so far to dispose of their refuse. In terms of the unusual characteristics of the Carpenter Brook artifact assemblage, Ritchie (1947:67) notes that typical early Late Woodland trash deposits

invariably produce only a scattered small fraction of the potsherds recovered at Carpenter Brook and a vastly larger representation than was found here of chipped and polished stone artifacts, plus bone and antler implements (always in the majority in this culture [i.e. the Canandaigua Focus] and here totally wanting), together with a far greater quantity of refuse bone, and rejectage of industrial processes. Moreover, no other instance of extensive massed sherds is known.

On the basis of these atypical qualities, Ritchie concluded that Carpenter Brook was the result of prehistoric ritual activity, a conjecture he developed even before completing his excavations; he refers to the site in his field notes as “a ceremonial dump” (Ritchie 1946). Interestingly, Mann (1922–1946) arrived at a similar conclusion, remarking in his notes that  “there seems to be some pertinent reason for this deposit ... of ceremonial origin.” Neither Ritchie nor Mann indicate whether they had discussed this possibility with one another. Ritchie's hypothesis concerning the nature of the rituals during which Carpenter Brook was deposited were mostly influenced by two of its characteristics: the large volume of ceramic material and the predominance of bear remains in the faunal assemblage. His ideas were also related to Hallowell’s (1926) influential study of bear ceremonialism, in which that author described a set of broadly similar ritual acts associated with bears throughout the northern hemisphere. Specifically, he noted practices in which some parts from bears that were consumed during rituals—typically their heads—were disposed of at ‘special’ places away from settlements. The bear remains were also sometimes left with small sacrifices of food, tobacco, or other items. Ritchie (1947:67, 71) suggests Carpenter Brook was the result of an analogous series of acts related to the ritual disposal of the remains from bears consumed during ceremony along with accompanying food offerings, “the spectacular breakage of pots being perhaps only incidental to this intrinsic purpose.” He also implies that the dramatic volume of pottery could potentially and inaccurately influence the interpretation of the events during which it was deposited: “Although the fictile component forms the dominant element in our discovery it may only mask the intrinsic features vital to the actual elucidation of the site” (Ritchie 1947:72).

In the remainder of this section, as well as the attribute analyses of the Carpenter Brook pots that follow it, I focus on an elaboration of Ritchie's ideas concerning the ceremonial origins of the site that considers its ceramic vessels as more central to the acts during which it was deposited than he believed. I then discuss another interpretation of the site that explores the possibility that it was related to ceramic production.

An alternative to Ritchie’s interpretation of Carpenter Brook is a scenario in which the pots from the site were not as tangential to the acts during which they were deposited as he believed. This is consistent with both ethnohistoric and archaeological data indicating Iroquoian-speaking people viewed ceramic vessels as symbolically charged items worthy of sacrifice. At the most general level, as Wonderly (2001) has noted, early historic-period Iroquois perceived strong symbolic associations between clay pots and fertility, the earth, and death and dying. While ceramic vessels in general would probably all have embodied these kinds of symbolic relationships to some degree, the associations would likely have been most intense when pots were directly involved in feasts, particularly those with ritual reenactments/reification of beliefs related to fertility, death, and dying. For example, the Huron referred to the ceremony popularly known as the “feast of the dead,” during which recently deceased individuals were interred in group burials, as ‘the feast of the kettle’ or more succinctly, “the kettle” (Thwaites 1896–1901, 10:269).1 Kapches (1976:33–34) reports on a particularly vivid archaeological manifestation of the link between cooking vessels and death, fertility and the earth from late prehistoric village sites in southeastern Ontario, where people interred deceased infants inside clay pots. The symbolic importance of cooking vessels was regularly reified by their central roles in Iroquoian social gatherings, including formal rituals described by Sagard and the Jesuits (e.g., Sagard 1939:211–212; Thwaites 1896–1901, 10:145, 179–180, 269–271, 289; 39:31; 70:149). Another indication of the symbolic prominence of pots was their role in metaphors related to general social conditions and well-being. For instance, the Jesuit de Brébeuf (Thwaites 1896–1901, 10:307) notes that one Huron group used the phrase “divided kettle” to describe a disagreement among villages and he notes that, in one case, “a general Assembly of the Notables of the whole Country took place to … reunite the kettle.” The Jesuits Chaumonot and Dablon report that both the Huron and Onondaga kept a “war-kettle” on the fire during times of conflict (Thwaites 1896–1901, 42:121–123, 169–171).

Archaeologically, the symbolic importance of ceramic vessels is suggested by their inclusion in Late Prehistoric and Protohistoric burials (Engelbrecht 2003:61; Tuck 1978:332; see, e.g., Ritchie 1937; Sempowski and Saunders 2001:732–808; Wintemberg 1936:117; Wray et al. 1987:175, 226). There is also some evidence for prehistoric instances of “pot killing” in the Northeast. For example, Jamieson (1999:191–192, note 3) and J. V. Wright (1999:683) have both suggested that clay pots from the southern Ontario Woodland period Pergentile and Red Horse Lake Portage sites, respectively, had been symbolically destroyed. Also, Ritchie (1944:227–228) has documented ‘killed’ steatite vessels accompanying Transitional Archaic period graves on eastern Long Island, each of which had a small hole knocked through its base. In light of this ethnohistoric and archaeological evidence for the symbolic importance of cooking vessels in the worldviews of Iroquoian-speaking people in the New York area, it is likely that those who visited Carpenter Brook and left items there had similar ideas about their clay pots.

Finally, from a purely pragmatic perspective, it would not have been productive for the individuals who visited Carpenter Brook to have simply smashed or left usable vessels behind if such an act had no symbolic purpose. Allen (1992:140–142) and Dunford (2001:38–59) have noted that prehistoric pottery production in the Northeast, which was most likely carried out by women, was a complex task that included gathering resources, forming vessels, and drying and firing them (also see Rice 1987:113–167). Many elements of this process were only possible in relatively dry and warm weather, so pottery production was likely limited to warmer months and was probably closely tied to the timing of other seasonal activities in which women participated, such as harvesting crops (Allen 1992:140–142; Dunford 2001:38–59). Undoubtedly, unexpectedly inclement weather, such as extended wet periods, would have contributed additional unpredictable elements to pottery production.

Thus, given the complexity and challenges of pottery production in the Northeast, as well the symbolic importance of ceramic vessels for Iroquoian-speaking people, it is unlikely that the pots from Carpenter Brook were “incidental” to the acts during which it was deposited, as Ritchie suggested. However, the possibility that the vessels from the site played a central symbolic role in ceremonies there is contingent on whether Ritchie’s ideas concerning the ritual nature of the site are accurate. His arguments relating the site to ceremonial acts were largely based on its atypical qualities, including the prominence of bear remains in the faunal assemblage, the large volume of pottery juxtaposed with the near-complete absence of lithic material, and its distance from the nearest contemporaneous habitation site (Ritchie 1947). However, his contention is consistent with additional ethnohistoric evidence related to the importance of liminal watery settings similar to those at the site, and ritual acts people performed at them (see Smith 2005).

In brief, Carpenter Brook is partially fed by an adjacent spring. In Iroquoian cosmology, springs and waterfalls are portals to a dangerous underwater world. There are numerous accounts from seventeenth-century European travelers in the Northeast including Sagard, the Jesuits, and Champlain, in which they report individuals leaving material sacrifices—including smoking pipes, tobacco, copper, and arrows—at such dangerous places in the landscape to ensure safe journeys (e.g., Champlain 2000:47; Sagard 1939:171, 189; Thwaites 1896–1901,10:159, 165-167; 50:265, 267, 287). J. V. Wright (1999:683–685) argued that pottery ranging from Middle Woodland to Late Prehistoric times found underwater at the Red Horse Lake Portage site in southeastern Ontario was left during a series of similar rituals (see also P. Wright 1980). In this light, Carpenter Brook also has some qualities commensurate with those of a ‘special place’ in the landscape at which people would have left items as offerings. It is somewhat distant from the nearest habitation site and is near a spring, an important and dangerous setting in Iroquoian world view. Beyond this, its artifact assemblage primarily comprises symbolically charged items, notably the pottery and bear remains—although the presence of less numerous objects, such as the smoking pipe[s], phallic effigy, and dog and puma remains, are probably also significant (Engelbrecht 2003:45–46, 54–57; Hamell 1998:269–271; Ritchie and Funk 1973:360; Smith 2005:112–130).

This evidence is consistent with Ritchie’s interpretation that Carpenter Brook was deposited during acts related to ritual. However, its ceramic vessels should not, as he argued, be dismissed as tangential to those acts and it is probable that they were intended as offerings as much as the items found among them. Thus, the pots from the site were used in a ritual social setting distinct from that of pots used in everyday/prosaic contexts. This interpretation permits analyses of the Carpenter Brook vessels from the perspective that some of their attributes might reflect the differing needs of the social setting in which they were used, relative to those of vessels found in more prosaic contexts.

The ethnohistoric and archaeological records contribute evidence to support the idea that the qualities of Iroquoian vessels changed with the social setting in which people used them. Most of these relate to feasting, one of the primary ways in which Iroquoian people shared ritual experiences in the early historic period (e.g., Thwaites 1896–1901, 10:177). For instance, in 1637, the Jesuit le Mercier recounted that a Huron individual reported to him that a healing ritual could not be performed because “they had no kettle large enough to make a feast” (Thwaites 1896–1901, 13:233). Beyond illustrating the pragmatic need for larger pots to cook for the greater numbers of people who attended a ceremonial feast, this individual’s statement indicates that several smaller vessels would not do—that is, that a large pot was symbolically necessary for conducting a feasting ritual. Elsewhere, Dunford (2001:124) has suggested that Late Woodland potters on Cape Cod produced two classes of vessels, one of which comprised “elaborately decorated” pots “for use in community-wide feasts” and the other was made up of “carefully constructed but minimally decorated…vessels for daily household use.” Meanwhile, Cervone (1987:24–25) and Wonderly (2002:38) have both raised the possibility that people manufactured vessels included in burials specifically for that purpose and Cervone has suggested that they might have characteristics different from pots they used everyday.


As indicated by le Mercier’s account, as well as the simple need for large vessels during feasting that could service more people than would a pot used for everyday cooking, a vessel’s size is one attribute that correlates, to some degree, with the social setting in which its maker intended it to be used (see also Mills 1999:104). Larger pots were likely meant, at least in part, to be used for feeding larger numbers of people, such as groups that would gather during ritual feasts. This is not to say that bigger vessels were not used for everyday purposes, but that in a ritual context that involved feasting one would expect cooking vessels to be larger. Thus, in the case of Carpenter Brook, the presence of relatively sizable pots would be another line of evidence consistent with the hypothesis that the site represents the remains of items used and consumed in a ritual context.

One variable that correlates with overall vessel size for prehistoric pots in the Northeast is orifice diameter. Thus, to examine the relationship between the sizes of the pots from Carpenter Brook and those used for everyday purposes, I compared the extrapolated orifice diameters of vessels from the site with those of pots from the early Late Woodland Bates, Maxon-Derby, and Sackett habitation sites. All three of these settlements were excavated by Ritchie at various times from the 1930s to the 1950s and their collections are stored at the Rochester Museum and Science Center (part of the Sackett assemblage) and the New York State Museum (Bates, Maxon Derby, and the remainder of the Sackett material) (Ritchie 1937; 1980:281–287; Ritchie and Funk 1973:195–252). Bates is roughly 100 km southeast of Carpenter Brook and has yielded radiocarbon dates from the calibrated twelfth to thirteenth centuries A.D. (Hart 2000; Ritchie and Funk 1973:251). Although the dates from Bates are 200 to 300 years later than those from Carpenter Brook, its two earliest dates (calibrated 2σ ranges and median probabilities of A.D. 1022 [1160] 1257 and A.D. 1019 [1161] 1276, respectively [Hart 2000:5]) are not significantly different from the later of the two dates for Carpenter Brook at the 95-percent level of confidence (Stuiver and Reimer 2010). Maxon-Derby, as mentioned above, is about 2 km south of Carpenter Brook and had multiple occupations: one during the calibrated eleventh century A.D. and another from the cal. mid-twelfth and to mid-thirteenth centuries A.D (Hart 2000). The two radiocarbon dates that represent the earlier of these occupations (calibrated 2σ ranges and median probabilities of A.D. 904 [1021] 1161 and A.D. 988 [1025] 1160, respectively [Hart 2000:5]), are not significantly different from the later of the two Carpenter Brook AMS dates at the 95-percent level of confidence, and the earliest date from Maxon-Derby is not significantly different from the earlier of the Carpenter Brook dates at the same level of certainty (Stuiver and Reimer 2010). Finally, the Sackett site is about 75 km west of Carpenter Brook. It has yielded radiocarbon dates from the calibrated thirteenth century A.D., roughly 300 years after Carpenter Brook (Hart 2000). Although all three of the village sites are within three centuries of the dates for Carpenter Brook and some of their dates are not significantly different than those from the site, there is a possibility that temporal differences among them are reflected in the results of the ceramic analyses presented below, an issue to which I return later. Also, since radiometric dates are not available for all the vessels included in this study, the attribute data presented below is necessarily an aggregate of values for each site. That is, it is not possible with current methods to determine, for example, when during the habitation of Bates a particular vessel was manufactured or to which occupation of Maxon-Derby a given pot may be assigned (with the exception of those sherds from contexts directly dated by Ritchie and Funk [1973] or Hart [2000]). This being said, the assemblages excavated from the village sites largely comprise items people used in “everyday” (not formal ritual) contexts. Thus the ranges of attributes of ceramic vessels from village settings will, to some degree, correspond with the prosaic functions for which the pots were intended (Allen 1992:139–140; Rice 1987:293–301). Undoubtedly, some sherds from vessels used for feasting will be present in village assemblages (e.g., Allen 1992:139–140), but the pots will likely be low in number relative to those used in everyday contexts.

Rim sherds from a total of 330 pots from the four sites were large enough to yield data for vessel diameter (Table 2.1; Figure 2.7). Of these, 114 are from Carpenter Brook and 216 are from the villages. The Carpenter Brook vessels have the largest average oral diameters, at 29.7 cm, compared with 21.1 cm for the village pots. The nonparametric Wilcoxon rank sum test indicates the distribution of diameter values for Carpenter Brook is significantly different from those of any of villages (α = .01); the same relation holds when the values for the village sites are grouped together (statistical calculations were performed with SPSS 9.0). Thus, the Carpenter Brook vessels are significantly larger (on average) than those from more prosaic contexts, a result that is consistent with the hypothesis that at least some of the pots from the Carpenter Brook site were used during feasting. Although more research is needed concerning how vessel diameters changed through time in central New York, it is unlikely the differences between the Carpenter Brook vessel diameters and those from the later village site pots reflect temporal changes. If, as suggested above, larger vessels were made to cook for larger numbers of individuals, one might also expect pot size to correlate with population density. Village sites with more than one household, among the earliest of which is Sackett, appear in central New York in the thirteenth century A.D (Hart 2000; Hart and Brumbach 2003:745–746). Thus, to some degree, populations were becoming denser during the time span represented by Carpenter Brook and the settlement sites. However, the earlier Carpenter Brook pots are larger than the pots from the later habitation sites, the opposite trend one would expect if the differences reflected increases in population density.

Table 2.1

Mean Reconstructed Vessel Orifice Diameters for Carpenter Brook and the Village Sites.

Figure 2.7

Frequency distribution of vessel orifice diameters from Carpenter Brook, Bates, Maxon-Derby, and Sackett.

Another attribute that may be related to the intended function of the Carpenter Brook pots is vessel wall thickness. In general, pots with relatively thin walls and consistently thick cross-section will be more durable in thermally stressful environments than will those with thicker walls or walls with inconsistent thickness (e.g. Rice 1987:227–229). In the Midwest, Braun (e.g. 1987:164) has noted a relationship between thinner-walled vessels and an increased dietary reliance on starchy seeds, which are best prepared by lengthy episodes of boiling or simmering. Brumbach and Bender (2002:235–236), Chilton (e.g., 1999:55, 58), and Schulenberg (2002b:88) have suggested a similar correlation in the Northeast (see also Luedtke 1986). While pots with thin walls of consistent cross-section are advantageous in terms of their long-term survivability when used repeatedly for cooking, they are also more time-consuming and require more skill to produce than pots with thick walls that have inconsistent thickness. Thus the qualities of the Carpenter Brook vessel walls relative to those of pots from the villages will shed some light on the degree to which the vessels were intended for long-term use for cooking.

Measurement of the thicknesses of 295 sherds from Carpenter Brook and 583 from the villages indicated that those from the site have the highest mean thickness, 7.97 mm; the average for the villages combined is 6.74 mm (Table 2.2, Figure 2.8) (for rim sherds, thickness measurements were taken at the furthest intact point from the vessel lip; for body sherds, the thickness measurement used was the average of the thickest and thinnest on the sherd). The Wilcoxon rank sum statistic indicates the Carpenter Brook distribution is significantly different from that of any of the village sites individually or when the village site data are pooled (α = .01). However, vessel wall thickness cannot be considered in isolation since, as noted above, the pots from the brook are also significantly larger than those from any of the other sites, suggesting its relatively high wall thickness value might be a reflection of its bigger vessels.

Table 2.2

Mean vessel wall thicknesses for Carpenter Brook and the village sites

Figure 2.8

Frequency distribution of vessel wall thickness values for Carpenter Brook and the villages.

The nonparametric Spearman’s rank correlation coefficient was employed to calculate the degree of correlation between vessel diameter and wall thickness for vessels from the four sites. Fragments from 112 pots from Carpenter Brook and 213 from the village sites (all from discrete pots) yielded data for both variables. Spearman’s coefficient indicates the two variables are correlated at the 95-percent confidence level for the four sites and for the data from the villages when they are grouped together. However, when the relationships between the data are smoothed using the LOWESS non-linear modeling method (after Hart and Brumbach 2009), there are two notable differences between the distribution from Carpenter Brook and that from the village sites (LOWESS smoothing was performed using the PTS LOWESS Calculator add-in for Microsoft Excel using a smoothing parameter of 1) (Figure 2.9). The trend line for the villages shows wall thickness and diameter are directly proportional for vessels smaller than about 40 cm in diameter: small pots tend to have thinner walls and larger ones have thicker walls. For pots from the villages larger than 40 cm. in diameter, wall thickness remains roughly the same, on average, regardless of vessel size (Hart and Brumbach [2009:373, 375] obtained similar results for data from Woodland-period habitation sites). For pots from Carpenter Brook, however, the trend line indicates a directly proportional relationship between the two variables for vessels of any diameter. There is no value for diameter above which pots tend to be made with walls of the same thickness. The other notable difference in the two trend lines is that, for the entire range of vessel diameter, the Carpenter Brook pots tend to have thicker walls. Thus, for vessels of a given diameter, those from the Carpenter Brook site generally were made with thicker walls relative to those from the villages. Both these aspects of the distribution of data for vessel diameter versus wall thickness are consistent with the idea that the potters who made its vessels were not as concerned with making thin-walled pots as were those who produced the vessels from the villages. This perhaps also implies that they were aware their wares would be employed as sacrifices, so they expended minimal effort toward qualities that would have assured the vessels had longer use-lives, but were more difficult to produce.

Figure 2.9

Scatterplot of vessel diameter vs. wall thickness for Carpenter Brook and the villages, also showing LOWESS smoothing.

It is likely that a portion of the differences in the thicknesses of the earlier Carpenter Brook pots and the later vessels from the villages is a reflection of a more general trend through time. Hart and Brumbach (2009:373–374) have shown there is a gradual decrease in vessel wall thickness throughout the Woodland period, from an average of over 11 mm at 1100 B.C. to around 7 mm after A.D. 1000—a total difference of about 4 mm over a period of 2,000 years (there may be a greater rate of decrease between A.D. 820 and 1130, but this depends on what technique is used for constructing trend lines with which to summarize the data). Thus, although this gradual trend may account for a part of the difference between the average value for the Carpenter Brook walls and those from the villages, it probably does not account for the entirety of the change which, when the villages are grouped, represents over a 1 mm decrease in thickness in a period of about 300 years. More research on how vessel wall thicknesses changed through time will undoubtedly shed more light on this issue.

Several sherds from Carpenter Brook display another quality related to wall thickness that indicates they were from vessels that would not have been durable as cooking pots and is also consistent with the hypothesis that their makers were expending minimal effort in assuring they had long use-lives. Specifically, they have gouges and depressions on their interior surfaces that were formed when their clay was still plastic—features that result in varying wall thickness across relatively small portions of the pots (Figure 2.10). If these pots were used repeatedly for cooking they would likely have soon crumbled from stresses caused by differential thermal expansion. No instances of similar internal gouges and depressions were observed on any of the pots from the villages.

Figure 2.10

Carpenter Brook sherds with irregular interior surfaces.

Finally, one Carpenter Brook sherd displays an anomalous quality that would have also made it short-lived in thermally stressful environments. The surface of the fragment is covered with a layer of clay that includes temper, but is likely unfired and that obscures the original exterior, lip surface, and interior of the portion of the vessel from which it came (Figure 2.11). Although the exterior of the added clay surface is in poor condition, it appears to be decorated. There were no other instances of this kind of vessel alteration in either the Carpenter Brook assemblage or that from the villages. The added clay might have been related to a repair, but too little of the original pot remains to permit a more definitive assessment of its possible function. Nonetheless, the added material would have decreased the durability of the vessel in cooking contexts. Not only does it increase the wall thickness of the pot from which the sherd came, but because it is made from clay with a different density than that of the underlying vessel, the pot’s walls would have had variable thermal expansion characteristics. Both qualities indicate the vessel’s wall would have expanded inconsistently in a thermally stressful environment and likely would have failed relatively quickly. Although one cannot read too much into a single anomalous artifact such as this fragment, its inferior characteristics for cooking are consistent with the qualities represented by many of the other sherds at the site, as well as the hypothesis that its maker(s) (or in this case, those individuals who altered it) were minimally concerned with cooking performance.

Figure 2.11

Carpenter Brook sherd with an additional applied layer of clay obscuring the original vessel surface. The original exterior surface of the vessel is visible in the upper right portion of this sherd.

Decoration Size and Complexity

Decoration size and complexity are additional vessel attributes a potter may have altered depending on the social context in which she intended her ware to be used. Both are expressions of the amount of ‘decorative effort’ expended on a pot (see Braun 1991:381). The hypothesis that the amount of decorative effort expended on a vessel varied with the social context in which it was to be used is based on the premise that pots used during feasting had central and highly visible roles. In Renfrew’s (1994:51) terminology, they served as “attention focusing devices.” Thus, larger and more complex decoration might have had the effect of enhancing the perceived relative visibility of a vessel during feasting, one of the most ubiquitous elements of Iroquoian ritual. Additionally, larger decoration would have been visible from greater distances and so increases in its size would also increase the number of individuals who could see it in crowded ritual/feasting settings (Carr 1995:189–190; see Rieth and Horton 2010:11–13). Finally, larger and more complex decoration might have been related to the greater diversity of people who attended feasts relative to those typically present in more prosaic contexts. Braun (1991:369–384) notes that the amount of decorative effort potters expended on their wares increased in southern Illinois between 200 B.C. and A.D. 200, a time when people from different kin groups started living together in composite households in increasingly diverse villages. In this context, ceramic vessels offered potters “a rich opportunity for decorative display and variation” (Braun 1991:384). Thus, although not ‘signaling’ (see, e.g., Wobst 1977) any kind of group affiliation, increases in the size and complexity of clay vessel decoration accompanied the intensification of inter-group communication in the Midwest. Similar (although briefer) phenomena may have accompanied ritual events in the Northeast during the Late Woodland, such as that represented by Carpenter Brook. Specifically, potters may have increased the amounts of decorative effort they expended on their wares in proportion to the diversity of individuals attending the events during which they were to be used.

Typically, decoration on early Late Woodland clay pots from New York is confined to the upper portions of vessels. It is most frequently applied with cord-wrapped sticks or other implements to the exterior of vessels’ necks extending down from their rims (the pots usually do not have collars). Decorative impressions also appear on pot lips and the upper parts of interior surfaces (also extending down from the rim). The following discussion focuses on exterior decoration, which is usually composed of bands of single motifs, including vertical, horizontal, and oblique impressions, as well as more complex zoned (i.e. plats) and chevron designs. Occasionally, potters also applied bands of punctuates or notches to the interior surface of the vessel (see, e.g., Ritchie and MacNeish 1949).

One way the amount of decoration on the exterior of a pot can be measured is in terms of the distance its decorative field extends below its lip. Rim sherds from a total of 44 vessels from Carpenter Brook and 56 from the villages are intact enough so this variable—the ‘height’ of the vessel’s exterior decoration—can be measured (Table 2.3, Figure 2.12). On average, the pots from Carpenter Brook have the greatest value for decoration height, 61.1 mm, larger than the values from any of the villages. However, the Wilcoxon rank sum test indicates that, at the 95-percent level of certainty, the distribution of values from Carpenter Brook is only significantly different from that of Sackett (α = .02). It is different from those from Bates and Maxon-Derby at the 72-percent and 83-percent levels of certainty, respectively. The Carpenter Brook distribution is also significantly different from those of the villages when the latter are grouped together, at the 95-percent level of certainty (α = .03). Thus, while the degree of statistical significance changes to some degree depending on how the data are grouped, the Carpenter Brook vessels do have larger decorative fields, on average, than do the pots from the villages.

Table 2.3

Mean Exterior Decoration Height and Average Number of Exterior Decorative Bands for Carpenter Brook and Village Vessels.

Figure 2.12

Frequency distribution of decoration height for the Carpenter Brook and village vessels.

This result might be linked to the fact that the pots are larger than those from the villages, suggesting there may be no proportional difference between the relative sizes of the decoration on the pots from the sites when vessel diameter is taken into account. That is, larger pots might simply have proportionally larger decoration. Spearman’s rank correlation coefficient was used to determine whether the two variables were related. Sherds from just 84 vessels yielded data for both vessel diameter and height of decoration; 41 are from Carpenter Brook and 43 from the villages. The results indicate there is no statistically significant correlation between the variables (even at just a 40-percent level of confidence) for vessels from Carpenter Brook (rs = -.035) or from the villages when their data are grouped together (rs = .079) (the data from the villages were pooled because of the relatively small numbers of sherds from each site that yielded data for both variables). Thus, it is unlikely that the larger decoration on the Carpenter Brook pots is related to their larger average size.

Design complexity, another indicator of decorative effort, can be measured as the number of distinct impressed bands composing a vessel’s decorative field. Based on the same vessels used for the ‘height of decoration’ analysis, the Carpenter Brook pots have the most decorative bands, on average: 3.2 per vessel (Table 2.3). The distribution of values for Carpenter Brook is only significantly different at the 95-percent level of certainty from that for Sackett (1.2 bands per vessel). The Carpenter Brook distribution is also significantly different at the 95-percent confidence level from those from the village sites when their data are grouped, but this result is heavily influenced by the low value from Sackett. In sum, the pots from Carpenter Brook do have more complex decoration in terms of the number of bands in their decorative fields, but the value is only slightly higher than that for Bates and Maxon-Derby. For this variable, Sackett, with its very low average, is the outlier.

Finally, although the average value for the number of decorative bands—i.e. design complexity—for the Carpenter Brook pots is only slightly higher than that for the villages, there may be differences between the two groups of pots in how the complexity of decoration on vessels is related to the size of their decorative fields. Spearman’s coefficient indicates there are statistically significant correlations between the two variables for all four sites and the villages when they are grouped. When the data from the sites (villages grouped) are smoothed with the LOWESS method (again using a smoothing parameter of 1), there are two notable characteristics in the trend lines (Figure 2.13). First, for vessels with relatively small decorative fields extending less than about 40 mm from vessel rims, those from Carpenter Brook and the villages tend to be decorated with similar numbers of bands per unit of decoration height. However, for vessels with larger decorative fields, those from Carpenter Brook typically have more complex decoration than do pots with similarly sized decorative field from the villages.

Figure 2.13

Scatterplot of size (‘height’) vs. complexity (number of bands) of the decoration on the Carpenter Brook and village vessels, also showing LOWESS smoothing.

In sum, the data support the interpretation that the potters who made the vessels recovered from Carpenter Brook expended more effort, on average, on the decorative qualities of their wares than did those who made the pots from the villages (although the degree of certainty of this assertion is certainly subject to interpretation based on how strictly one wishes to adhere to the statistical standard of utilizing a 95-percent level of confidence as a standard critical value for whether to reject null hypotheses in archaeological applications). The Carpenter Brook vessels have larger mean decorative fields than do those from the villages, and the difference is not related to the larger average size of its pots. The Carpenter Brook vessels also have more decorative bands—i.e. greater decorative complexity—than do the village pots, although the statistical significance of this result is somewhat equivocal. Finally, it appears that, for vessels with large decorative fields, the Carpenter Brook pots display more complex decoration than do vessels from the villages with equally sized decorative embellishment. More research is needed to explore how these decorative variables changed temporally in order to explore whether the differences discussed here are related to changes through time from the earlier Carpenter Brook to the later habitation sites.


One possible alternative to the hypothesis that Carpenter Brook was formed during ritual activities is that it was related to pottery production. Allen (1992:144) describes several archaeological indicators that may be present at or near loci of vessel manufacture, including: “pottery making tools, proximity to clay kilns, stashes of clay and temper, and a high number of wasters (broken during the process of manufacture).” In the Northeast, vessels were likely either fired in ovens or in open fires, which would have left archaeological "evidence in the form of areas of burned soil, possibly fire-reddened ... and associated with piles or scatters of sherds" (Allen 1992:144).

Several characteristics of Carpenter Brook correspond with elements of this description. These include the possible vessel-making tool found by Mann and the potential presence of clay near the site, which he also noted. Beyond this, the Carpenter Brook pots, which tend to have qualities that make them poor cooking vessels, may have been wasters, vessels disposed of during or immediately after production because they had manufacturing errors. This would also be in line with Mann's idea that the fragile nature of the pots was a result of them being poorly fired. Although neither Ritchie nor Mann noted the presence of burned or reddened soil, such features would have been easy to miss if buried.

However, the hypothesis that Carpenter Brook was related to pottery production cannot account for several of the site's other qualities, such as the presence of the anomalous faunal assemblage (the predominance of the bear remains) and the clay phallus, or the fact that some of the vessels from the site had been used repeatedly for cooking, as evidenced by the presence of adhered charred food residue. These qualities are more consistent with the hypothesis that the site was related to ritual activity. However, neither interpretation of the site is necessarily exclusive of the other. Although the AMS dates for the site are not statistically distinct, they suggest it may have been deposited over a period of several decades. This scenario would also be consistent with the large volume of material from the site. Thus, if Carpenter Brook was deposited over several decades, it is certainly within the realm of possibility that its uses alternated during that time and may have included those related to both pottery production and ritual practices. Beyond this, additional uses not considered here might also be feasible.


The attributes of the Carpenter Brook pots examined in this paper—diameter, wall thickness, and the size and complexity of decoration—are consistent with the interpretation that the potters who made the vessels intended at least some of them to be used for purposes different from those from the more prosaic contexts at Bates, Maxon-Derby, and Sackett, although variables such as changes through time or the possibility that Carpenter Brook was related to pottery production may also have influenced the results. The Carpenter Brook pots on average are larger than those from the villages, suggesting more of them were intended to feed large numbers of people, such as would be present during feasting. Their walls, on average, have qualities that would have made them less durable as cooking vessels than those from the villages; i.e. they would not have endured as many heating-cooling episodes. Finally, as an assemblage, they have larger and more complex decoration than do pots from the villages (even when the larger sizes of the pots from the brook are factored into the equation). Thus, while they have inferior qualities relative to prosaically practical long-term use as cooking vessels, their decoration indicates their makers expended more effort on their embellishment than they did on pots that were used in “everyday” social contexts. Finally, as Mann noted, some of the Carpenter Brook sherds show minimal indications of use, such as adhered charred cooking residue, and their fragile nature may be attributed to a poor or minimal firing process. All of these characteristics are consistent with the hypothesis that the pots from the site were not intended for long-term use as cooking vessels.

The symbolically charged qualities of Carpenter Brook, including its setting, which is similar to those at which people were observed to make sacrifices of objects in historic times, and other elements of its artifact assemblage, such as the bear remains and smoking pipes, along with the fact that ceramic vessels themselves were items worthy of being sacrificed, all imply that the pots from the site were intended as symbolic offerings. Thus, the differences between them and vessels from more prosaic contexts might reflect the differing needs of those two social settings. The Carpenter Brook pots, if their makers intended them to be used as offerings, would not have needed the difficult-to-produce qualities that would have ensured they could withstand the thermally stressful environment of repeated use as cooking pots. It is conceivable they were made with just enough thermal durability to survive a single use. At the same time, their larger sizes would have served the greater numbers of people who would be present during a ritual event than would pots used in everyday contexts. Their larger and more complex decoration may have served as attention-focusing devices, important elements of feasts during which food-containing vessels would have been at the center of the ritual performance. The larger average sizes of the Carpenter Brook pots may also have served attention-focusing roles. These assertions are tempered by the possibility that Carpenter Brook may also be related to ceramic vessel production and some of its vessels might be wasters—pots that were recognized as of low quality during the manufacturing process.

Undoubtedly, there are numerous other potential interpretations for the disparities between the Carpenter Brook pottery and the vessels from the village sites beyond those addressed in the current study. Also, since the analyses presented here have focused on comparing the Carpenter Brook pots as an assemblage with those from the villages, the variability among its vessels has been downplayed. For example, while the pots from the brook are large on average and many have thick walls that have been minimally refined, small vessels with thin, finely made walls are also present. Also, while most of the Carpenter Brook pots show few indications of long-term use such as adhered charred cooking residue, a small number of sherds (approximately five) do have adhered residue (Mann’s observation that the sherds from the site appeared to have been minimally used indicates the lack of residue is not a reflection of post-excavation cleaning). Further study focusing on the refined pots or those with charred residue would undoubtedly be productive. There are also attributes of the Carpenter Brook vessels relative to those from the villages that were not addressed in the current study, such as temper. An analysis of the qualities of the temper of the Carpenter Brook pots relative to those from the other sites might shed further light on the differing needs of the purposes for which the two groups of vessels were intended. Additionally, an analysis that includes some earlier domestic sites would elucidate how potential changes in ceramic vessel attributes through time have influenced the results presented here. Finally, a more systematic survey of the area around Carpenter Brook than that conducted by Ritchie might provide additional evidence concerning other site uses, including its possible role in pottery production.

Although this study has focused on the clay vessels from Carpenter Brook, the remainder of its assemblage also holds potential for additional research. For example, the clay phallic effigy might represent an opportunity for studying gender roles, particularly in the context of a site with so many ceramic vessels. In early historic times, clay pipes were apparently made by men (Engelbrecht 2003:55; Kuhn 1996:32), and the physical form of the phallus from Carpenter Brook unmistakably conveys a sense of masculinity. Clay pots, meanwhile, were made by women (e.g. Allen 1992:140; Sagard 1939:109) and were associated with fertility. The proximity of two such explicit expressions of gender at a possible location of repeated ritual acts represents an additional potentially productive subject for future study. Beyond this, residue analysis of material that may be present in the perforation of the phallic effigy might more definitively address whether the item was used for a smoking pipe. Also, given the socially integrative function of ritual in general, it is even conceivable that men played a role in manufacturing the vessels from the site. This would be consistent with the presence of vessels whose more difficult-to-produce attributes were poorly executed, since men were probably fairly inexperienced in making pottery. However, testing this hypothesis with archaeological data would be very challenging.

The perspective that the differences between the Carpenter Brook pots and those from the villages are related to the distinct social settings in which the vessels’ makers intended them to be used adds another vector of variability for pottery analyses in the Northeast. It also sheds some light on the roles of individuals relative to long-duration technological changes, such as the gradual decrease in vessel wall thickness through the Late Woodland noted by Hart and Brumbach (2009). The differences in the wall thicknesses of the vessels from Carpenter Brook relative to those from the village sites indicate that the potters who made them were aware of the importance of controlling such subtle variables and, in so doing, allude to the degree to which individuals played roles in the gradual changes in ceramic technology through the Woodland Period (Hart and Brumbach 2009:368–369). This is also consistent with Braun’s (1983:112) observation that numerous ethnographic studies of traditional pottery manufacture from contexts around the world indicate that potters are aware of the subtlest attributes of their wares. Finally, it underscores the fact that a site of a type as singular as Carpenter Brook can contribute to broader-scale investigations of social, economic, and cultural dynamics in the Northeast, despite its uncommon characteristics.


Thanks to Christina Rieth and John Hart for editing this volume and organizing the NYSAA session upon which it is based. Also, thanks go to Bill Engelbrecht for his encouragement and feedback. I am grateful to Jim Bradley for recommending Frederick Dunford’s dissertation and for kindly supplying a copy of Earl Mann’s Carpenter Brook notes. Thanks also to Gian Carlo Cervone and Kathryn Murano at the Rochester Museum and Science Center and Andrea Lain at the New York State Museum, all of whom provided a great deal of friendly assistance as I collected the data for this project, as well as to the three anonymous reviewers who provided numerous useful comments on an earlier draft of this paper. Finally, I am grateful for the financial support for the project, provided by the Robert E. Funk Memorial Archaeology Foundation, which generously supplied funds for the Carpenter Brook AMS dates, and NSF IGERT Award DGE-9870668.


1Much of the ethnohistoric evidence cited in the discussion that follows of beliefs related to Carpenter Brook are based on observations made by Champlain, Sagard and the Jesuits among the Huron in the seventeenth century. Although the Huron were an Iroquoian-speaking group and likely had many customs and beliefs similar to those of Iroquoian-speaking people who inhabited New York State in prehistoric times, there were undoubtedly many differences as well. As Allen (1992:135) has noted, “the written accounts can only be used as guides to the earlier situation ... However [they] may be considered a baseline for an understanding of Iroquoian society.” These issues—particularly the degree to which practices and beliefs recorded in written historical accounts can be used for interpreting Carpenter Brook, which dates to about six centuries before the earliest interactions between native groups in the New York area and Europeans—are addressed further in Smith 2005:71–225.

References Cited

Allen, K. M. S. 1992. Iroquois Ceramic Production: A Case of Household-Level Organization. In Ceramic Production and Distribution: An Integrated Approach, edited by G. J. Bey III and C. A. Pool, pp. 133–154. Westview Press, Boulder.

Braun, D. P. 1983. Pots as Tools. In Archaeological Hammers and Theories, edited by James A. Moore and Arthur S. Keene, pp. 107–134. Academic Press, New York.

Braun, D. P. 1987. Coevolution of Sedentism, Pottery Technology, and Horticulture in the Central Midwest, 200 BC–AD 600. In Emergent Horticultural Economies of the Eastern Woodlands, edited by W. F. Keegan, pp. 153–181. Center for Archaeological Investigations, Occasional Paper No. 7. Southern Illinois University, Carbondale.

Braun, D. P. 1991. Why Decorate a Pot? Midwestern Household Pottery, 200 B.C.–A.D. 600. Journal of Anthropological Archaeology 10:360–397.

Brumbach, H. J., and S. Bender. 2002. Woodland Period Settlement and Subsistence Change in the Upper Hudson River Valley. In Northeast Subsistence Settlement Change AD 700–1300, edited by J. P. Hart and C. B. Rieth, pp. 227–239. New York State Museum Bulletin 496. The University of the State of New York, Albany.

Carr, C. 1995. A Unified Middle-Range Theory of Artifact Design. In Style, Society, and Person: Archaeological and Ethnological Perspectives, edited by C. Carr and J. E. Neitzel, pp. 171–258. Plenum, New York.

Cervone, G. C. 1987. Seneca Pottery Analysis: Some Problems and Solutions in Refining the Potential of Attribute Analysis. The Bulletin: Journal of the New York State Archaeological Association 95:23–31.

Champlain, S. de. 2000 [1604-1616]. Algonquians, Hurons, and Iroquois: Champlain Explores America, 16031616. Translated by A. N. Bourne, edited by E. G. Bourne. Brook House Press, Dartmouth, Nova Scotia.

Chilton, E. S. 1999. One Size Fits All: Typology and Alternatives for Ceramic Research. In Material Meanings: Critical Approaches to the Interpretation of Material Culture, edited by E. S. Chilton, pp. 44–60. University of Utah Press, Salt Lake City.

Dunford, F. J. 2001. Ceramic Style and the Late Woodland Period (1000-400 B.P.) Sachemships of Cape Cod, Massachusetts. Unpublished Ph.D. dissertation, Department of Anthropology, University of Massachusetts, Amherst.

Engelbrecht, W. 2003. Iroquoia: The Development of a Native World. Syracuse University Press, Syracuse, New York.

Gates St-Pierre, C. 2001. Two Sites, But Two Phases? Revisiting Kipp Island and Hunter’s Home. Northeast Anthropology 62:31–53.

Hallowell, A. I. 1926. Bear Ceremonialism in the Northern Hemisphere. American Anthropologist 28(1):1–175.

Hamell, G. R. 1998. Long-Tail: The Panther in Huron-Wyandot and Seneca Myth, Ritual, and Material Culture. In Icons of Power: Feline Symbolism in the Americas, edited by N. J. Saunders, pp. 258–291. Routledge, New York.

Hart, J. P. 1999. Dating Roundtop’s Domesticates: Implications for Northeast Late Prehistory. In Current Northeast Paleoethnobotany, edited by J. P. Hart, pp. 47–68. New York State Museum Bulletin 494. The University of the State of New York, Albany.

Hart, J. P. 2000. New Dates from Classic New York Sites: Just How Old are Those Longhouses? Northeast Anthropology 60:1–22.

Hart, J. P., and H. J. Brumbach. 2003 The Death of Owasco. American Antiquity 68:737–752.

Hart, J. P., and H. J. Brumbach. 2005. Cooking Residues, AMS Dates, and the Middle-to-Late Woodland Transition in Central New York. Northeast Anthropology 69:1–33.

Hart, J. P., and H. J. Brumbach. 2009. On Pottery Change and Northern Iroquoian Origins: An Assessment from the Finger Lakes Region of Central New York. Journal of Anthropological Archaeology 28:367–381.

Hart, J. P., R. G. Thompson, and H. J. Brumbach. 2003. Phytolith Evidence for Early Maize (Zea mays) in the Northern Finger Lakes Region of New York. American Antiquity 68:619–640.

Hutton, F. Z., and C. E. Rice. 1977. Soil Survey of Onondaga County, New York. United States Department of Agriculture, Soil Conservation Service, U.S. Government Printing Office, Washington, D.C.

Jamieson, S. M. 1999. A Brief History of Aboriginal Social Interactions in Southern Ontario and Their Taxonomic Implications. In Taming the Taxonomy: Toward a New Understanding of Great Lakes Archaeology, edited by R. F. Williamson, and C. F. Watts, pp. 175–192. eastendbooks, Toronto.

Kapches, M. 1976. The Interment of Infants of the Ontario Iroquois. Ontario Archaeology 27:29–39.

Kuhn, R. D. 1996. A Comparison of Mohawk and Onondaga Projectile Point Assemblages. Journal of Middle Atlantic Archaeology 12:27–34.

Luedtke, B. E. 1986. Regional Variation in Massachusetts Ceramics. North American Archaeologist 7:113–135.

Mann, E. 1922–1946. “Large Deposit of Third Period Algonkian Earthenware in the East Bank of Carpenter’s Brook.” Notes on file at the New York State Museum, The University of the State of New York, Albany.

Mills, B. J. 1999. Ceramics and the Social Contexts of Food Consumption in the Northern Southwest. In Pottery and People: A Dynamic Interaction, edited by J. M. Skibo, and G. M. Feinman, pp. 99–114. University of Utah Press, Salt Lake City.

Parker, A. C. 1922. The Archaeological History of New York, Part I. New York State Museum Bulletin 235, 236. The University of the State of New York, Albany.

Prezzano, S. C. 1988. Spatial Analysis of Post Mold Patterns at the Sackett Site, Ontario County, New York. Man in the Northeast 35:27–45.

Reimer, P.J., M.G.L. Baillie, E. Bard, A. Bayliss, J.W. Beck, P.G. Blackwell, C. Bronk Ramsey, C.E. Buck, G.S. Burr, R.L. Edwards, M. Friedrich, P.M. Grootes, T.P. Guilderson, I. Hajdas, T.J. Heaton, A.G. Hogg, K.A. Hughen, K.F. Kaiser, B. Kromer, F.G. McCormac, S.W. Manning, R.W. Reimer, D.A. Richards, J.R. Southon, S. Talamo, C.S.M. Turney, J. van der Plicht, and C.E. Weyhenmeyer. 2009. IntCal09 and Marine09 Radiocarbon Age Calibration Curves, 0–50,000 Years Cal BP. Radiocarbon 51(4):1111–1150.

Renfrew, C. 1994. The Archaeology of Religion. In The Ancient Mind: Elements of Cognitive Archaeology, edited by C. Renfrew, and E. Zubrow, pp. 47–54. Cambridge University Press, New York.

Rice, P. M. 1987. Pottery Analysis: A Sourcebook. University of Chicago Press, Chicago.

Rieth, C. B., and E. Horton. 2010. Stylistic and Technological Analyses of Ceramic Vessels from the Bailey Site, Onondaga County, New York. In Preserving Tradition and Understanding the Past: Papers from the Conference on Iroquois Research, 2001–2005, edited by C. S. Patrick, pp. 5–14. New York State Museum Record 1. The University of the State of New York, Albany. Electronic Document,, accessed November 23, 2010.

Ritchie, W. A. 1937 [1936]. A Prehistoric Fortified Village Site at Canandaigua, Ontario County, New York. Researches and Transactions of the New York State Archaeological Association, Lewis H. Morgan Chapter, Vol. VIII, No. 2. Lewis H. Morgan Chapter, Rochester, New York. Reprinted from: Research Records of the Rochester Museum of Arts and Sciences, No. 3. Rochester Museum of Arts and Sciences, Rochester, New York.

Ritchie, W. A. 1944. The Pre-Iroquoian Occupations of New York State. Rochester Museum Memoir No. 1. Rochester Museum of Arts and Sciences, Rochester, New York.

Ritchie, W. A. 1946. “Carpenter Brook Site: Ceremonial Pottery Dump.” Notes on file at the Rochester Museum and Science Center, Rochester, New York.

Ritchie, W. A. 1947. Archaeological Evidence for Ceremonialism in the Owasco Culture. Researches and Transactions of the New York State Archaeological Association, Lewis H. Morgan Chapter, Vol. XI, No. 2. Lewis H. Morgan Chapter, Rochester, New York.

Ritchie, W. A. 1980. The Archaeology of New York State. Rev. ed. Purple Mountain Press, Ltd., Fleischmanns, New York.

Ritchie, W. A., and R. E. Funk. 1973. Aboriginal Settlement Patterns in the Northeast. New York State Museum Memoir 20. The University of the State of New York, Albany.

Ritchie, W. A., and R. S. MacNeish. 1949. The Pre-Iroquoian Pottery of New York State. American Antiquity 15(2):97–124.

Sagard, G. 1939 [1632]. The Long Journey to the Country of the Hurons. Edited by George M. Wrong, Translated by H. H. Langton. Champlain Society Publication XXV. Champlain Society, Toronto.

Schulenberg, J. K. 2002a. New Dates for Owasco Pots. In Northeast Subsistence-Settlement Change AD 7001300, edited by J. P. Hart, and C. B. Rieth, pp. 153–165. New York State Museum Bulletin 496. The University of the State of New York, Albany.

Schulenberg, J. K. 2002b The Point Peninsula to Owasco Transition in Central New York. Unpublished Ph.D. dissertation, Department of Anthropology, Pennsylvania State University, State College.

Sempowski, M. L., and L. P. Saunders. 2001. Dutch Hollow and Factory Hollow: The Advent of Dutch Trade Among the Seneca, Parts I-III. Research Records No. 24. Charles F. Wray Series in Seneca Archaeology, Volume III. Rochester Museum and Science Center, Rochester, New York.

Smith, D. A. 2005. Carpenter Brook Revisited: Social Context and Early Late Woodland Ceramic Variation in Central New York State. Unpublished Ph.D. dissertation, Department of Anthropology, University at Buffalo, The State University of New York, Buffalo.

Soil Survey Staff. 2011. Web Soil Survey of Onondaga County, New York. Natural Resources Conservation Service, United States Department of Agriculture, Web Soil Survey. Web-based Data Exchange System and Electronic Document,, accessed May 8, 2011.

Stuiver, M., and P.J. Reimer. 1993. Extended 14C Database and Revised CALIB Radiocarbon Calibration Program. Radiocarbon 35:215-230.

Stuiver, M., and P.J. Reimer. 2010. CALIB 6.0.1. Program and Electronic Document,, accessed August 23, 2010.

Thwaites, R. G. (Editor). 1896–1901. The Jesuits Relations and Allied Documents: Travels and Explorations of the Jesuit Missionaries in New France, 16101791; the Original French, Latin, and Italian Texts, with English Translations and Notes. 73 vols. Burrows Brothers, Cleveland.

Tuck, J. A. 1978. Northern Iroquoian Prehistory. In Northeast, edited by Bruce G. Trigger, pp. 322–333. Handbook of North American Indians, Vol. 15, William C. Sturtevant, General Editor, Smithsonian Institution, Washington, D.C.

Wintemberg, W. J. 1936. Roebuck Prehistoric Village Site, Grenville County, Ontario. National Museum of Canada, Bulletin No. 83. Anthropological Series, No. 19, Ottawa.

Wobst, H. M. 1977. Stylistic Behavior and Information Exchange. In For the Director: Research Essays in Honor of James B. Griffin, edited by C. E. Cleland, pp. 317–342. Anthropological Papers, Museum of Anthropology, No. 61. University of Michigan, Ann Arbor.

Wonderly, A. 2001. Oneida Pottery Effigies. Paper Presented at The 68th Annual Meeting of the Eastern States Archaeological Federation, Watertown, New York.

Wonderly, A. 2002. Oneida Ceramic Effigies: A Question of Meaning. Northeast Anthropology 63:23–48.

Wray, C. F., M. L. Sempowski, L. P. Saunders, and G. C. Cervone. 1987. The Adams and Culbertson Sites. Research Records No. 19. Charles F. Wray Series in Seneca Archaeology, Volume I. Rochester Museum and Science Center, Rochester.

Wright, J. V. 1999. A History of the Native People of Canada, Volume II, 1,000 B.C.–A.D. 500. Mercury Series. Archaeological Survey of Canada Paper 152. Canadian Museum of Civilization, Hull, Quebec.

Wright, P. J. 1980. Prehistoric Ceramics from the Red Horse Lake Portage Site (BdGa-12) Eastern Ontario. Archaeology of Eastern North America 8:53–70.

« previous   |   next chapter »

Museum Open Tuesday-Sunday: 9:30 am to 5 pm | Carousel Hours: 10 am to 4:30 pm
Office of Cultural Education | New York State Education Department
Information: 518-474-5877 | Contact Us | Image Requests | Terms of Use
Join us on Facebook See us on YouTube See us on Flickr