Archive for May, 2011

Kiva R, Pueblo Bonito

I’ve been talking quite a bit lately about the idea that kivas in Chacoan great houses were residential spaces, but it’s important to note that there are in fact other rooms at these sites that show at least as much evidence for residential use as the kivas do.  These are typically large square or rectangular rooms facing plazas, often with T-shaped doors opening into the plazas, and they have firepits, storage pits, mealing bins, and other features typically interpreted as indicating residential use.  Arguments that Chacoan great houses were not used residentially at all tend to gloss over the presence of these rooms, and arguments that great houses may have had some residential functions but were primarily used for other purposes tend to focus on the small numbers of these rooms at excavated sites such as Pueblo Bonito and Chetro Ketl.

And, indeed, not very many rooms have these features.  Using the fantastic database that the Chaco Archive is putting together, I find only 134 “thermal features” at Pueblo Bonito out of about 400 excavated rooms.  Note that this figure includes all formal hearths as well as more ephemeral heating pits, and that it includes kivas as well as rectangular rooms.  I do think it’s reasonable to think that most kivas were residential, of course, so this isn’t as big an overestimate as it might seem at first glance, but it’s still an overestimate.  Storage features are even less prevalent, with only 107 documented at Pueblo Bonito.  Furthermore, chronological control during early excavations was not great, so it’s not totally clear when in the site’s long occupation history these rooms were used for residential purposes; it’s quite possible that they started out as residential rooms and later were converted to some other use when the use of the whole site changed.

Western Burial Rooms in Old Bonito

Still, those plaza-facing square rooms do pose a challenge for Steve Lekson‘s arguments that kivas were residential, because they seem to take the place of the kivas he posits within residential room suites.  Indeed, some even have kiva-like features like ventilation shafts, particularly in the western part of “Old Bonito” where many of the rooms were later used as burial chambers.  Tom Windes has pointed to this as a challenge to Lekson’s interpretation of kivas, and indeed it seems tricky to interpret.

Why are there two different types of rooms that both seem to indicate similar residential uses?  Were kivas and plaza-facing square rooms part of the same room suites?  Were the kivas sunk into the plaza in front of the room suites used for some sort of multi-household purpose, as Windes has proposed, rather than being individual household residential spaces?  If so, what?  It is noteworthy that, to the extent that we can tell based on the existing architecture, there generally seem to have been too few kivas in early great houses for each room suite to have had its own.  It’s also striking that while some room suites at early great houses are very obvious sets of interconnected rooms, others are much more difficult to interpret and may not have had quite the same functions.  And it’s here that I think a possible answer to this puzzle may emerge.

Kiva T, Pueblo Bonito

Looking at the distributions of kivas and square rooms with residential features, I think it’s possible that there were two separate traditions or styles of domestic architecture that we see in early great houses: one based on a “Prudden unit” with a kiva and a few surface rooms used for storage, and another based on a “room suite” with a large living room, containing a hearth, taking the place of the kiva and being directly connected to two or three smaller storage rooms behind it.  Where these styles may have come from and what, if anything, they represented socially is still an open question as far as I’m concerned, but it’s something I’ll be looking into.  Tentatively, I’m thinking regional variation in architecture is perhaps the most likely answer, especially given other lines of evidence suggesting that Chaco was a multiethnic community incorporating people from a variety of geographic and cultural backgrounds.  It’s also possible, however, that these differences reflect some other dimension of social diversity, or that they have no relation to any such type of diversity that we can see in the archaeological record today.

If the people who lived in kivas and the people who lived in square rooms were different people in some meaningful sense, that has potentially important implications for many aspects of the Chaco system.  I’ll try to tease out these implications in upcoming posts, and I’ll also look at some other lines of evidence supporting this idea.

Partly Walled-Up T-Shaped Doorway between Room 28 and the Plaza, Pueblo Bonito

Read Full Post »

Unit Pueblo with Single Kiva at Mesa Verde

My kiva posts so far have mostly focused on form rather than function, but function is really at the core of most arguments over kivas these days.  As I said earlier, I basically buy Steve Lekson‘s arguments that small (as opposed to great) kivas in the San Juan region before AD 1300 were basically residential in function and formed integral parts of household residential units consisting of a kiva and a few rectangular rooms.

Identifying these units at small houses, which constitute the vast majority of sites in the eleventh century and earlier, is very straightforward.  Many of these sites include just one kiva and a few rectangular rooms.  These are the famous “Prudden units” first identified in the Mesa Verde region by T. Mitchell Prudden a century ago.  Each seems to be pretty clearly a residential structure housing a single household.  Other small sites include two or more Prudden units, but most have no more than a few and seem to indicate the presence of multiple households, probably extended families.

Cliff Palace, Mesa Verde

At the large aggregated sites that began to develop in the twelfth century and reached their greatest size in the late thirteenth century, just before the abandonment of the whole region during the “Great Drought” of 1276 to 1299, the most famous of which are the cliff dwellings at Mesa Verde such as Cliff Palace and Spruce Tree House, it is generally possible to identify individual suites of rooms with associated kivas equivalent to Prudden units.  This suggests that these sites formed by the aggregation of existing social units without major changes in social organization.  After the major societal dislocations and migrations in the period between 1275 and 1325, this is no longer the case, and the much larger aggregated pueblos of the Rio Grande Valley, as well as those in the Hopi and Zuni areas to the west, no longer have well-defined units with associated kivas.  Instead they have many fewer kivas, indicating a new function for these spaces more like their current function at modern pueblos as community-wide ritual and integrative spaces.  This function is probably more like the function of great kivas in earlier periods, although it is important to note that post-1300 community-scale kivas do not continue the formal attributes of previous great kivas, Chacoan or otherwise, and instead more closely resemble the smaller kivas that Lekson considers residential rather than ceremonial in function.  They are, however, generally quite a bit larger than typical pre-1300 small kivas.  It is also in this era that we begin to see the distinction in kiva form between eastern and western pueblos that is continued to the present day, with eastern pueblos adopting round kivas and western pueblos adopting square ones.

The idea, then, assuming Lekson is right, is that in the course of the massive changes around 1300 people adopted new social organizational principles, one of which was the abandonment of the kiva as a residential space and the development of a new ceremonial and social space, shaped like an earlier residential kiva but larger, that may have eased people’s integration into the new communities that were much larger and denser than what they had known before.  Increased density would have led to increased social tensions, and new social institutions, probably including the kachina cult, would have been necessary to deal with these tensions.  Because of all these changes, the modern pueblos are probably not good analogues for understanding pueblo life before 1300, even though they are clearly and unambiguously descended from the people who occupied pueblo sites in such areas as Chaco and Mesa Verde at that time.

Chetro Ketl from Above

So that’s all well and good, but notice what’s missing from this story: the great houses associated with the system centered on Chaco between 1030 and 1130.  Most discussion of Chacoan architecture has focused on these very impressive buildings, particularly on the largest ones, Pueblo Bonito and Chetro Ketl, even though the vast majority of sites even at Chaco Canyon itself were small houses, and Lekson’s edited volume on Chaco architecture is no exception.  Recently one of the major debates in Chacoan archaeology has focused on the function of these structures, with some arguing that they were primarily ceremonial or administrative in function with minimal residential use, while others see them more as residences of an elite class in a hierarchical society (Lekson has referred to them as “palaces”) and a few still arguing that they were residential communities analogous to the later aggregated pueblos.  Sorting this out has been tricky, especially since some of the most important sites in Chaco were excavated long ago when archaeological techniques were not what they are now and figuring out the implications of data collected at that time is very difficult.  There is clearly evidence for at least some residential use at most great houses that have been excavated, but it remains unclear how much of this evidence dates to the period of Chacoan florescence rather than earlier or later, and a number of more recent lines of evidence, such as the extensive road system throughout the San Juan Basin associated with the great houses and the elaborate landscape architecture and astronomical alignments found in the vicinity of many great houses, has led others to argue that the primary function of these structures was more esoteric and ceremonial, even if a few people did live in them.

One challenge has been determining the residential population of these sites.  Indeed, if it were easy to determine population many of the disputes over function would probably go away.  Early estimates based on raw room counts and comparisons with modern pueblos led to estimates as high as 1200 people for Pueblo Bonito alone, but more recent estimates based on the actual presence of residential features in (rectangular) rooms have gone as low as 70 for Pueblo Bonito.   That’s quite a range!

Pueblo Bonito from Above

So what does this have to do with kivas?  Well, if kivas were residential, then obviously they would have relevance to debates over residential use and population estimates.  If, as seems to have been the case at small sites, each kiva corresponds to one household unit, a rough count of population could come from just counting up the number of kivas in use at a given point in time and applying some conversion factor based on the assumed size of a household.  As far as I know no one has actually done this, although Lekson or someone else may have done it somewhere I haven’t seen.

So, assuming kivas were residential, and noting the numbers of kivas at great houses, we can perhaps get a rough sense of the number of people who lived in these buildings.  But who were these people?  And why did they live in great houses while most other people in their communities, both inside and outside Chaco Canyon, lived in the much humbler small houses?  The obvious answer is that they were the elites in a hierarchical system, but that poses as many questions as it answers.  What was the nature of this system?  Where did the elites get their authority, and how did they maintain it?

Kiva F at Pueblo Bonito

Setting aside these more fundamental questions about Chaco, however, there are also many questions raised by the idea of kivas being residential that focus more on the kivas themselves.  As I mentioned before, the classic “Chacoan” kivas found at the great houses have a very standardized set of features that distinguishes them from kivas elsewhere in the region.  Were these characteristics of elite residences?  If so, understanding their purpose would potentially be very useful in understanding the dynamics of the Chaco system more generally.  Unfortunately, we don’t really have any idea what these specialized features may have meant socially or economically.  They don’t seem particularly resource-intensive compared to other kiva styles, which suggests that conspicuous consumption may not have been the motivation.  This makes sense, since if kivas were private, residential spaces few people would presumably have had access to them.

It’s also possible that the specialized features of Chacoan kivas were markers of ethnicity or heritage.  In general they seem to differ most noticeably from those found in areas to the north, and they may resemble southern forms more closely.  There are other suggestions of southern origins for at least some early great house architecture which may be instructive, and great kivas, which became integrated with great houses about the time the Chaco system really got going, are also generally more of a southern phenomenon.  This idea of regionalism is important, I think, but I don’t have enough information to discuss it more fully just yet.

One of the most striking differences between Chacoan kivas and those found elsewhere in the region is the tendency for Chacoan ones to be elevated and contained within square rooms, whereas most others were subterranean.  Is this an “elite” feature?  It certainly would make these residences more prominent and given their inhabitants a better view of whatever was happening in the plazas of the great houses.  It would also have increased the visibility of these residences from the clifftops above the great houses in Chaco, which is discussed in the chapter on Chetro Ketl in Lekson’s architecture volume.  Also discussed in that chapter is the Court Kiva at Chetro Ketl, one of the small number of exceptionally large “Chacoan” kivas built into plazas rather than elevated in roomblocks.  The Court Kiva was sited well into the plaza, where it was separated from the nearest rectangular rooms by a greater distance than was typical but was also particularly visible from the cliff above the site.  It was later converted into a (presumably non-residential) great kiva, which is interesting and unknown anywhere else at Chaco.

Elevated Kiva G at Chetro Ketl

This same chapter does contain some discussion of the topic of kiva function at great houses which is worth quoting.  The differing interpretations of Lekson and Windes are here made explicit; it is not uncommon for coauthors of scholarly publications to disagree about some aspects of how to interpret their research, but it is only in publications on Chacoan archaeology (and this is not the only one I’ve seen do this) that they actively seek to dissociate themselves from each other’s interpretations.  Here’s Lekson’s take:

Lekson believes “kivas” at small sites were residences and, at great houses, elite residences—conspicuous and expensive housing.  Building big, flashy houses would have posed a problem for emerging elites at Chaco: If homes are traditionally below grade, how do you make one conspicuous?  Elevate it, get it out of the ground and up where it can be seen.  The Kiva G complex, ending with a third-story “kiva,” is a remarkable example.  But even elevated, enclosed kivas are not obvious from ground level; the viewer sees the rectangular exterior walls, not the house form itself.

And here, in the very next paragraph, is Windes:

Windes does not discount the possibility that “kivas” were used in part as habitations, but he does not believe that they were “living rooms” in the traditional sense.  Rectangular, surface living rooms are in fact found in great houses, but given the differences in location, shape, floor features, and artifacts, we must ask what roles these two very different types of structures took.  In addition, Windes believes that the small kivas represent part of the triad of structures associated with basic domestic units, as found in small sites and tucked away in the corners of great houses during the last occupations, but that the very large “court” kivas, later shifted in location from the plazas to the roomblocks, are specific to Chaco great houses and represent some activity beyond mere domestic use; in fact, their very size, sometimes with the feature attributes of great kivas, would place them as great kivas anywhere else outside the confines of the great house.

Both of these interpretations are plausible, but I still find Lekson more convincing.  What is the “activity beyond mere domestic use” that Windes sees these kivas serving?  He doesn’t specify.  He does, however, point out another important aspect of kivas that I alluded to in the previous post: the proliferation of them late in the occupation periods of many great houses, often associated with rectangular (or irregularly shaped) surface rooms and looking very similar to Prudden units or equivalent single-household residential roomblocks.  The masonry on these tends to look very late, and they are most often found sunk into late plaza surfaces, so while they can rarely be dated directly it seems likely that they represent a subsequent reuse of the sites as locations for residential settlement, probably after they were no longer in use for their original purposes, whatever those might have been.  Many of the original square rooms in some great houses seem to have been used for trash deposition at this time, implying that they were no longer needed for their original functions but that there were plenty of nearby residents producing trash.  Some great kivas, however, especially Kiva A at Pueblo Bonito, also seem to have continued in use at this point, implying that the sites may still have had some ritual or integrative functions, even if they were newly being used for a new type of residential use.

Kiva A at Pueblo Bonito

What Windes doesn’t mention here, however, is that many of the late “non-Chacoan” kivas found at great houses were built into existing square roomblocks, the same way earlier Chacoan kivas had been built into newly constructed square rooms.  This implies more continuity in both architecture and population than his phrasing would suggest between the large “court kivas” and the later ones.  Recall that Lekson has also argued that these non-Chacoan kivas are more likely a late development of Chacoan architectural traditions than a sign of an influx of “foreigners.”

Windes is certainly right that there are square rooms with residential features in Chacoan great houses, and that this suggests that not everyonewas necessarily living in kivas, but the features of many of those square rooms are actually pretty similar in many cases to kiva features.  This implies that the distinction between those who lived in kivas and those who lived in aboveground square rooms that look suspiciously like kivas may be more complicated than Windes (or Lekson) lets on.  Perhaps there is some social distinction here, implicit in the architecture, that has received little to no attention in the literature but that may well be quite important.  But that’s a topic for another post.

Room 330, Pueblo Bonito

Read Full Post »

Late Kivas in the Southeast Corner of the Chetro Ketl Plaza

Implicit in my previous discussion of “Chacoan” kivas was the idea that the term “Chacoan” in this context refers to a specific architectural form defined by a collection of features, rather than to a geographic location.  Thus, Chacoan kivas are common at Chaco Canyon, but they are also found at many sites outside the canyon, particularly at Chacoan “outliers” or sites with great houses similar to those at Chaco and other attributes that tie them to Chaco despite quite considerable distances.  The converse is also true, in that not all kivas at Chaco are Chacoan kivas.

Kivas in the Southeast Part of Pueblo Bonito

Since the Chacoan kiva form is so standardized and consistent, it is fairly easy to tell when a given kiva does not meet the criteria to be considered Chacoan, and many kivas at Chaco do not.  In his 2007 chapter on great house form, Steve Lekson identifies 22 excavated kivas at Chaco great houses that do not meet Chacoan criteria.  Since these are considerably smaller than the excavated Chacoan kivas, with an average diameter of 4.3 meters (as compared to an average diameter of 7.2 meters for Chacoan kivas), Lekson refers to them as “small round rooms.”  This category originated as something of a catch-all for kivas that did not meet all the criteria to be considered Chacoan kivas, great kivas, or tower kivas, but in addition to the small size there are some other commonalities among these rooms.  None has the full set of Chacoan kiva features, but many do have some of these features, including subfloor ventilators, southern bench recesses, and floor vaults.  However, only three have bench recesses, and of those only two also have subfloor ventilators.  These two may actually be Chacoan kivas, although they are missing the floor vaults and beam pilasters standard to the Chacoan type, and Lekson speculates that it may be their small size that leads them to lack these features.  They were both built into elevated square rooms, which is typical of Chacoan kivas.  Lekson defines his categories based on size rather than internal features because he wants to include unexcavated rooms, which is reasonable, but it does seem likely that these two kivas at least really are the smallest Chacoan kivas and fit the type criteria imperfectly because of their size.

Keyhole-Shaped Kiva at Pueblo Bonito

Other non-Chacoan kivas are more obviously different in form.  One of the most conspicuous non-Chacoan attributes of many kivas is that they have high masonry pilasters rather than low ones with radial beams; another is that they have a “keyhole” shape rather than being circular.  This shape results from a southern recess not just of the bench, as in Chacoan kivas, but of the walls as well, and it is quite common in the Mesa Verde region to the north, as is the use of high masonry pilasters to support a cribbed roof.  In that area the keyhole shape is often associated with a ventilation shaft that opens horizontally into the recess, as described by Jesse Walter Fewkes for kivas at Spruce Tree House, but almost all of the non-Chacoan kivas at Chaco have the “Chacoan” subfloor vent type instead.  There are only two excavated examples that do not have subfloor vents, both at Pueblo Alto; one of these has a Chacoan-style bench recess and the other has a floor vault.  In other words, no excavated kiva at a Chaco great house is completely lacking “Chacoan” features.  Lekson uses this fact to argue that even these non-Chacoan kivas are more likely “local” than “foreign” (as they have often been considered), and he concludes “many, perhaps most, of the smaller round rooms represent a late expression of the Chacoan buildings tradition.”

Kiva E at Pueblo Bonito, with High Masonry Pilasters

Foreign or not, they certainly do seem to be late.  Most were added into existing square rooms in a fashion similar to that seen with elevated, blocked-in Chacoan kivas, but it is important to note that Chacoan elevated kivas were usually added into rooms that were specially built for the purpose, while non-Chacoan ones were typically added into square rooms that had previously had other uses.  None of these kivas produced tree-ring dates, but Lekson considers them all to probably date to the early twelfth century or later.  Subterranean examples, generally built into plazas, also seem to be late.  The ones in the southeast corner of Chetro Ketl, which are still exposed today and can easily be seen, are associated with a late plaza surface.  Similar ones at Pueblo Bonito have mostly been backfilled completely and are no longer visible, but they probably date to the same period.

Keyhole-Shaped Kiva, Aztec Ruins National Monument

Given the increase in cultural influence from the north, especially from Aztec Ruins and the Totah area around modern-day Farmington, New Mexico, in the early 1100s, it makes sense that later Chacoan architecture would start to show northern influences during this period even if there were not a major influx of immigrants from the north (for which there isn’t really any evidence).  I therefore think Lekson’s arguments for a local origin for these features are reasonable.

Kiva W at Pueblo Bonito

Many of the currently visible kivas at Chaco, both elevated in roomblocks and subterranean in plazas, are of this type.  This has important implications for understanding what you are looking at when you visit Pueblo Bonito and Chetro Ketl.  Not all of the kivas you see were occupied or used at the same time, and many of them were later additions that were probably not part of the original plan for the areas where they are located.  Furthermore, there are lots of other kivas that you can’t see, mostly under the plaza.  To the extent that what you see at these sites reflects a moment in time, that moment was probably very late, perhaps even after the decline of Chaco from its heyday around AD 1100.  Much of what you see exposed today may not have been visible then, however, and there was a lot more that you can’t see that probably was.  I’m not going to go into the question of the function of all these kivas and the implications of that for the function of the sites containing them right now, although it is an important question, but the idea of change over time is important and I just want to emphasize it in a general sense right now.
Fewkes, J. (1908). Ventilators in Ceremonial Rooms of Pre Historic Cliff-Dwellings American Anthropologist, 10 (3), 387-398 DOI: 10.1525/aa.1908.10.3.02a00020

Read Full Post »

Entrance to Kiva at Edge of the Cedars State Park, Blanding, Utah

Sticking with the topic of the small round rooms traditionally called “kivas,” which Steve Lekson would prefer to call simply “round rooms,” it’s important to note that there is a wide variety of formal types.  In addition to the modern distinction between square and round kivas, which is basically geographical with square ones in the western pueblos and round ones in the eastern pueblos, and setting aside the highly specialized “great kivas,” among the prehistoric kivas (I’m going to stick with the traditional term for now) of the San Juan Basin there are at least two types.  In his writings on Chacoan architecture, Lekson has distinguished between two main types of kivas found in great houses at Chaco: “Chacoan” and “Non-Chacoan.”

Kiva Z, Pueblo Bonito

The type of kiva that Lekson defines as “Chacoan” (originally defined by Neil Judd, who excavated Pueblo Bonito and Pueblo del Arroyo in the 1920s) has a variety of standard features, especially in the later examples from the period of approximately AD 1075 to 1130 when the Chaco system was at its height.  They are not quite as standardized as Chacoan great kivas, but the features associated with them are nevertheless found wherever there is evidence for Chacoan influence during this period, and it seems clear that this particular suite of features is a specifically Chacoan development.  (These kivas have often been called “clan kivas” in the past, but I don’t like that term because of the huge assumptions it makes about social organization and kiva function, so I’m just going to call them “Chacoan kivas.”)  The standard features defined by Judd are:

  1. A central firepit
  2. A subfloor ventilation system with an opening south of the firepit leading to a shaft opening south of the kiva
  3. A subfloor “vault” west of the firepit
  4. A bench around the circumference of the kiva
  5. 6 to 10 low “pilasters” roughly evenly spaced around the bench
  6. A shallow recess in the bench at the southern end

Lekson adds two more features, which are certainly present in many Chacoan kivas but less universal than Judd’s and more controversial:

  1. The elevation of the kiva into an aboveground square enclosure
  2. “Wainscoting” around the edge of the bench

This set of features is certainly consistent with the general “San Juan” type of kiva that developed out of the Basketmaker pithouse, but it differs from the kivas found most commonly in areas like Mesa Verde to the north in a few ways.  Before going into the differences, though, I want to just explain the importance of the lists of features given by Judd and Lekson.

Kiva Firepit at Lowry Pueblo in Colorado

Firepit: All kivas have firepits; it is one of the defining characteristics of the form.  In Chacoan kivas specifically, the firepit is offset slightly to the south of the center point of the kiva, which is always circular.  Firepits in Chacoan kivas are deep, circular or square in plan, and usually lined with masonry.

Subfloor Ventilation Shaft in Kiva at Edge of the Cedars State Park, Blanding, Utah

Ventilation System: One major characteristic of San Juan small kivas in general is that they have ventilation shafts, usually at the southern end.  Jesse Walter Fewkes wrote an article in 1908, which I mentioned in an earlier post, in which he set forth an argument that these shafts were indeed for ventilation rather than for any other purpose, and this argument is now more or less universally accepted.  There are different types of ventilation system, however, and this is one of the major features distinguishing Chacoan kivas from other types.  Chacoan kivas have ventilation shafts that run underneath the floor of the kiva and are accordingly called “subfloor” ventilation shafts.  One end of the shaft opens vertically into the floor just south of the firepit, and there may or may not be a slab or low wall in between used as a deflector to distribute the air and shelter the fire.  From this opening the shaft runs down a short distance then turns and runs horizontally to the south underneath the floor (or as a shallow trough that would have been covered by boards or poles) until it gets past the southern wall, at which point it turns again and runs vertically upward until it reaches the ground surface (at the level of the kiva roof, but just to the south of it) and opens up to provide the source for fresh air.

Great Kiva at Chetro Ketl Showing Floor Vault

Floor Vault: Most Chacoan kivas have a single rectangular “box” sunk into the floor just to the west of the firepit.  These are often filled and plastered-over, and sometimes have boards covering them, so Lekson notes that this feature may actually be more widespread than it appears from the literature (since excavators may have missed covered vaults in some cases).  Since about three-quarters of excavated Chacoan kivas had evidence of vaults, this suggestion implies that these may have been nearly or literally universal in actual fact.  These vaults are reminiscent of the similar “vaults” known from Chacoan great kivas, although its unclear why there would be different numbers of them.  In both great and small kivas the function of the vaults is obscure.  The fact that they sometimes have wooden boards on them has led some to argue that they were “foot drums” that people would have danced on to create a drumming sound, but Lekson points out that they are often filled with sand, which makes this explanation implausible.

Chacoan Kiva at Aztec Ruins National Monument

Bench: There is a low masonry bench around the circumference of the room.  This is another standard feature of San Juan kivas in general, although the bench is not always made of masonry in non-Chacoan versions.

Kiva Pilasters at Pueblo Del Arroyo

Pilasters: At roughly equal intervals around the bench there is a series of “pilasters.”  This term comes from Mesa Verde kivas where the pilasters are often tall and made of masonry, and it is not as applicable to Chacoan kivas where the defining feature of a “pilaster” is a short segment of a wooden log oriented radially with one end set in the wall just above the bench.  These beams are often set in small masonry cubes which do somewhat resemble Mesa Verdean pilasters and imply a similar function.  Mesa Verdean pilasters typically serve to support a cribbed roof, and Chacoan pilasters have often been interpreted similarly, although Lekson disagrees with this interpretation.  The issue of roofing is discussed more fully below under “wainscoting.”

Kiva I at Pueblo Bonito Showing Southern Recess

Recess: At the south end of the bench there is a shallow “recess” in which the bench narrows.  The location of the recess corresponds to the location of the subfloor vent shaft, but since the vent shaft is underground it does not actually have anything to do with the recess (this is another difference from Mesa Verdean kivas, which have above-floor vent shafts that open into the recess, which is often more prominent).  There is some evidence that at least in some cases there may have been a shelf over the recess, which would have continued the line of the bench and created a large niche under it.  The purpose of this recess is obscure.

Southern Recess in Kiva at Pueblo del Arroyo

Those are the criteria Judd gives, and they are pretty universally accepted and uncontroversial.  Lekson adds two more, which are a bit more controversial.

Corner of Room Containing Blocked-In Kiva at Tsin Kletzin

Elevation and Blocking-In: The early examples of Chacoan kivas at Chaco great houses, dating from around AD 900 to 1070, are generally subterranean and usually located in the plazas of great houses, backed by suites of rectangular rooms.  The “classic” examples of Chacoan kivas, dating from about 1075 to 1130, are generally built into square rooms within the great-house roomblocks, usually on the first floor but occasionally on the second.  Lekson considers this tendency to “block-in” kivas a key part of the Chacoan kiva tradition, and in his 2007 chapter on great house form he goes into some detail on the historical development of the Chacoan kiva, starting with the early tenth-century examples, which are poorly known, and continuing through what he refers to as “transitional” Chacoan kivas, built between 1030 and 1070, only a few of which have been excavated.  The best known of these is Kiva G-5 at Chetro Ketl, which was later covered over by later kiva construction culminating in an elevated “classic” Chacoan kiva (Kiva G) but is still kept open and visible underneath the later construction.  These transitional kivas had most of the characteristics of later elevated kivas, and by Judd’s standards they would all be considered just Chacoan kivas.  Lekson makes a big deal about the blocking-in, however, and it is true that this is something that markedly distinguishes Chacoan kivas from other types.  No one else did this, and it’s very odd in a structural sense since those huge masonry cylinders needed extensive support, which often meant the “interstitial” rooms in the corners of the square room were braced with timbers or filled in with earth.  One problem with using this as a defining characteristic of Chacoan kivas, though, is that there are a few late, very large Chacoan kivas that are subterranean and located in plazas rather than being blocked-in.  These approach great-kiva size, but they lack the features of great kivas.  The best known of these is the Court Kiva at Chetro Ketl, which was later remodeled into a great kiva.  Only two other examples have been excavated, Kiva R at Pueblo Bonito and Kiva J at the Talus Unit.  Kiva R has standard Chacoan kiva features, whereas Kiva J was only partially excavated and little is known about its features.  Five additional kivas like this are known at Pueblo Bonito, and Lekson describes them as unexcavated, although at least two or three of them clearly seem to have been excavated as far as I can tell and they seem to have typical Chacoan kiva features, so I’m not sure what Lekson’s talking about when he says they’re unexcavated.  Indeed, one of these, Kiva O, is still visible in the east plaza.  (Kiva R, which is in the west plaza, is also visible.)  The fact that some of the largest Chacoan kivas are subterranean and in the plazas of great houses rather than elevated and blocked in makes Lekson’s use of blocking-in as a standard attribute of Chacoan kivas problematic, even just looking at the “classic” Chacoan kivas built after 1075.

Kiva L, Pueblo Bonito

Wainscoting: This is the most controversial of Lekson’s criteria for Chacoan kiva status.  Basically, many of the excavated Chacoan kivas have a series of thin wooden poles (or, less often, boards) rising from the back of the bench and leaning in toward the center of the ceiling.  Between them is a sort of wickerwork held together with clay or adobe (i.e., a sort of wattle-and-daub or jacal), plastered with mud on the interior side.  The space behind this wickerwork is either left open or filled in with trash or other vegetal material (Lekson’s account is unclear here).  Lekson claims that this “wainscoting,” supported by the poles, formed the ceiling of the kiva, sort of a false dome, with the exterior roof at the top being supported by horizontal beams much like those used in the roofing of standard square rooms.  This is in contrast to the standard way that Mesa Verde kivas were roofed, which was also a false dome but one made of cribbed logs beginning on the pilasters and alternating rows up to the roof.  (This is the way Navajo hogans are traditionally roofed as well.)  Some examples of intact roofs like this are reported in the Mesa Verde region, including one at Square Tower House that Fewkes used as the basis for interpreting and reconstructing the roofs of kivas at Spruce Tree House, which had not survived intact.  There is at least one kiva at Pueblo Bonito that also had a largely intact cribbed roof (Kiva L).  It has often been assumed that most Chacoan kivas, including the blocked-in ones, also had cribbed roofs resting on the pilasters, but it’s noteworthy that Kiva L is not blocked-in, although it does otherwise show classic Chacoan features, and that Kiva 67, another plaza kiva with classic Chacoan features, also showed evidence of having a cribbed roof through the impression of a log in clay spanning two pilasters, although the log itself did not survive.  It’s possible, then, that the development of “wainscoting” as a means to roof kivas was an innovation spurred by the building of kivas in square rooms, which could easily be given flat roofs like other square rooms, although it’s not really clear what the advantage of wainscoting over cribbing would have been.  It would probably have used less timber, but the Chacoans were hardly averse to importing huge quantities of timber and it’s hard to see them making decisions about architecture based on efficient use of resources.  Chacoan kiva roofing remains an open question.

Cribbed Kiva Roof at Edge of the Cedars State Park, Blanding, Utah

Kivas are particularly vulnerable to deterioration if they are left open to the elements, so all of the small kivas at Chaco that have been excavated have been subsequently backfilled to varying degrees.  Many have been filled entirely, so that no trace of them remains on the surface; this is the case with the Court Kiva at Chetro Ketl and many of the plaza kivas at Pueblo Bonito.  Others have only been refilled partly, in some cases to a low level so that the bench and pilasters are still visible and in other cases to a higher level so that only the upper parts of the wall can be seen.  Thus, there is nowhere at Chaco where the floor features of a Chacoan kiva can be seen.  This is in contrast to Mesa Verde, where especially at the cliff dwellings like Spruce Tree House many well-preserved kivas in sheltered locations have their floors open to be examined.   Those are generally Mesa Verde-style kivas, of course, rather than Chacoan ones.  The best example I know of a basically Chacoan small kiva where the floor features can be seen is the reconstructed blocked-in kiva at Edge of the Cedars State Park in Blanding, Utah.  This is an outlying great house that is much more modest than what you see at Chaco, but one of its kivas has been given a restored cribbed roof and other reconstructed elements to give a sense of what it would have likely looked like in its prime, and as it happens this kiva shows most elements of the Chacoan style despite being far from Chaco itself and in the Mesa Verde region.  Also in the same region, one of the kivas at Lowry Pueblo has not been totally reconstructed to the same extent but it does have a protective roof over it and so also has its floor features open.  This is another blocked-in kiva at an outlier far to the north that is nonetheless a good example of classic Chacoan kiva design.
Fewkes, J. (1908). Ventilators in Ceremonial Rooms of Pre Historic Cliff-Dwellings American Anthropologist, 10 (3), 387-398 DOI: 10.1525/aa.1908.10.3.02a00020

Read Full Post »

Reconstructed Great Kiva, Aztec Ruins National Monument

Although the idea that the small round rooms that area so common at Chacoan sites are ceremonial “kivas” has been increasingly challenged recently, it is still widely accepted that the large, formal, round structures known as “great kivas” were in fact community-wide ceremonial or integrative facilities.  Even Steve Lekson agrees, and he continues to use the term “kiva” in referring to these structures even as he calls the small “kivas” “round rooms” instead.  (He also uses the term “kiva” in referring to “tower kivas,” yet another form of round structure with proposed ceremonial associations.)  Ruth Van Dyke‘s chapter in The Architecture of Chaco Canyon, New Mexico is a good summary of current knowledge about Chacoan great kivas.  The great kiva is an architectural form that predates Chaco, and it may or may not have outlasted it.  The Chacoan form is distinctive, however, and found even in areas without a long history of pre-Chacoan great kivas.  It is highly standardized in both size and features, and is one of the surest indications of Chacoan influence wherever it is found.

Floor Features of Kiva Q, Pueblo Bonito

The following features are always found at Chacoan great kivas, although their specific realization can vary a bit:

  1. Four post holes, arranged in a square, to support the beams or columns that hold up the roof.  The holes may be either round or square.  Generally the columns themselves would be huge wooden beams, stubs of which have sometimes been found in the post holes during excavation.  Sometimes, such as in the great kiva at Aztec Ruins, square masonry columns, possibly with small poles in them, would be used instead.  It’s apparently not totally clear if the use of square rather than round post holes necessarily indicates the use of pillars rather than beams, since the beams would typically be held in place by shale and this could be done in either a square or a circular space.  When beams were used, they were supported at the bottom by several stacked stone disks, presumably to distribute the weight.  Offerings of turquoise and other valuables were often found in the beam holes, apparently placed during construction.
  2. Around the circumference of the kiva is a bench, sometimes doubled.  These benches were often refaced with new masonry, sometimes in connection with more general renovation of the kiva features and sometimes not.
  3. There is typically a series of wall niches around the circumference of the chamber, above the bench.  These vary in dimensions and number, but there are usually about 30 of them, especially in later great kivas.  Sometimes there is more than one series of niches at different levels, as at Casa Rinconada.  The purpose of the niches is unclear; some of them had offerings sealed into them, but these may have been construction offerings rather than indicating anything about post-construction use.
  4. Entrance is from a staircase leading down from an antechamber.  There would probably have been a smokehole in the roof as well, but it is unclear whether there would have been a ladder providing entrance through the roof as was the case in smaller round rooms.  An intact great kiva roof has never been found, which is unsurprising since the roofs would have been enormously heavy and very likely to cave in once the structure was no longer maintained.  The antechamber is on the north side in most cases.  Kiva Q at Pueblo Bonito has an apparent staircase and antechamber on the south side instead, but Van Dyke suggests that this may have been an error of reconstruction.  She doesn’t go into any more detail about this, however, and it’s unclear what the implications are if the room on the south side of Kiva Q is not an antechamber.  Casa Rinconada has antechambers with staircases on both the north and south sides.
  5. Along the central north-south axis, slightly offset to the south from the center point, is a firebox.  This is usually a masonry cube with a circular or oval firepit in it.
  6. Just south of the firepit there is a deflector.  This is a common feature in small kivas, which usually have a ventilation shaft on the south side, but since great kivas don’t have ventilation shafts and usually have their entrances on the north side it is unclear how useful this deflector would have been in practice.  Assuming there was a smokehole, a great kiva was big enough that it’s unlikely ventilation would have been a major concern.
  7. Attached to the two southern postholes on the north side, and sometimes running all the way to the northern postholes, there are two rectangular masonry “vaults.”  They are usually but not always subterranean.  The function of these is unclear.  Some have claimed that they are “foot-drums,” which would have had boards on top of them and people dancing on them, but not everyone accepts this interpretation and I don’t find it very convincing.  Small kivas sometimes have a single subfloor vault on one side of the firepit, but it is unclear if there is any connection between that type of feature and the much more formal vaults of great kivas.

These are the basic features that are repeated again and again at Chacoan great kivas.  Relatively few have been excavated, but all of those that have show these same features with minor variations.  Van Dyke provides a comprehensive list of the known great kivas at Chaco.  There are 21 of them, of which 11 have been excavated.  Ten of these are associated with the great houses Pueblo Bonito (4 great kivas), Chetro Ketl (3), and Kin Nahasbas (3).  (Note that Van Dyke is counting remodeled versions of earlier great kivas separately here.)  The only “isolated” great kiva to be excavated is Casa Rinconada.  It is also the largest excavated great kiva in the canyon at 19.5 meters in diameter, although it is not the largest excavated great kiva (the one at Village of the Great Kivas, a Chacoan outlier on the Zuni Reservation, is 23.7 meters in diameter), nor is it the largest great kiva in the canyon (the unexcavated northwest great kiva at Peñasco Blanco is 23 meters in diameter).

Casa Rinconada, Looking North

Van Dyke explicitly cautions her readers to be careful about the possibility of overemphasizing the importance of Rinconada just because it is so well known, and this is an important warning.  It does appear that Rinconada is unusual among all known great kivas in several ways, including the two antechambers and the “secret tunnel” leading from a back room of the north antechamber to a subsurface round enclosure around the northwest posthole.  It is also positioned in a very significant location, across from Pueblo Bonito, and there may be astronomical alignments encoded into it.  However, it is important to note that like the other great kivas at Chaco that are visible today, Rinconada has been substantially reconstructed.  In general Chaco has had a much lighter touch with reconstruction than many other parks, but great kivas, which are typically found in a substantially reduced state with large v-shaped breaches in the upper walls, are an exception.  Kivas A and Q at Pueblo Bonito as well as Casa Rinconada have all been built up to what their excavators considered a reasonable approximation of their original condition.  The great kiva at Aztec, of course, has been completely reconstructed to give an impression of what it might have looked like, and while there was apparently once talk of doing something similar at Casa Rinconada nothing ultimately came of it.

In addition to the excavated great kivas, there are ten unexcavated ones at Chaco.  It is hard to tell much about these, since they are basically just big recessed circles in the ground, but they are generally at least in the same size range as the excavated examples and can probably be assumed to be similar.  There may well be additional unknown ones, either associated with great houses or isolated.  It is particularly likely that early great kivas would not be apparent on the ground, since they are generally smaller than later ones and the excavated examples (or possible examples) all come from within early great houses where they are often overlain by later construction.

Northwest Great Kiva at Peñasco Blanco

The known unexcavated great kivas associated with great houses include two at Una Vida, one at Hungo Pavi, and four at Peñasco Blanco.  There are also three “isolated” great kivas, all of them at the east end of the canyon: one in Fajada gap, one on the south side of the canyon across from Wijiji, and one in a side canyon at the foot of Chacra Mesa below the Basketmaker III village known as Shabik’eshchee.  As noted above, the northwest one at Peñasco Blanco is huge, probably the largest at Chaco.  The one in Fajada gap appears to be about 20 meters in diameter, which puts it in the same size range as Casa Rinconada, although the difficulty of measuring diameter precisely with unexcavated great kivas makes it impossible to say if it is actually bigger than Rinconada or not.  One interesting thing about these isolated great kivas is that they are all on the south side of the canyon, as is Casa Rinconada.  This contrasts with the tendency of great houses to be on the north side and provides some support for the idea that the great kiva is conceptually separate from the great house and has its own history as a form.  It’s hard to say how to interpret this in the context of the postulated attempt by great-house elites to incorporate great kivas into their great houses as a way to legitimize their authority, which Van Dyke proposes as an explanation for why most great kiva construction at great houses didn’t take place until the mid-1000s.

And, indeed, early great house construction does seem to be notably bereft of great kivas.  Or does it?  Tenth-century “great kivas” are in fact postulated at Pueblo Bonito, Una Vida, and Kin Nahasbas, and Van Dyke includes them on her list, but it is unclear whether they really “count” as great kivas.  They are smaller than the later versions, which may just be because they are older.  They are also poorly documented, however; the ones at Pueblo Bonito and Kin Nahasbas have been excavated, but records about them are scarce and scanty.  The one at Pueblo Bonito is about 10 meters in diameter, which Van Dyke considers “within the range known for domestic pitstructures,” and furthermore it lacks postholes for roof support beams but does appear to have pilasters on its bench, which implies a roofing system like that of small kivas.  Since the roofing system is one of the most consistent features of classic Chacoan great kivas, this is a major strike against great kiva status for this one.  However, it’s possible that the specialized roofing system for later great kivas was an innovation to handle the large size of the ones built from the mid-1000s on, and that earlier structures with “regular” kiva roofs may have had “great kiva” functions in the 900s.  (This reminds me that I should do a post on small-kiva roofing, which is an interesting and surprisingly contentious issue.)

Kin Nahasbas from Una Vida

Evidence that the specialized roofing system for great kivas was already in place in the 900s comes from the early “great kiva” at Kin Nahasbas, which was more thoroughly excavated than the one at Pueblo Bonito.  It underlies the two later great kivas, which had classic great kiva features.  Its own features were largely obscured by the later construction, but it does appear to have postholes.  It couldn’t be dated directly, but the excavators concluded that it was probably associated with the tenth-century greathouse behind it.  This implies that there was at least one great kiva this early, but that the one at Pueblo Bonito was not one.  Interestingly, the diameter of this great kiva was only 7 meters, making it smaller than the Pueblo Bonito example and suggesting that size isn’t everything when it comes to great kivas.

The early great kiva at Una Vida is very poorly known and may not exist at all.  There is certainly another, later great kiva at the site.  Van Dyke refers to William Gillespie’s account of Una Vida’s architecture in Steve Lekson’s Great Pueblo Architecture of Chaco Canyon, New Mexico as the source for the idea that there is a great kiva associated with the early-tenth-century construction there, but Gillespie is very vague about the basis for his speculation that such a great kiva existed, and says only that “surface evidence is inconclusive.”  Van Dyke lists the diameter of this postulated great kiva as 17 meters, which is remarkably large for such an early structure and only slightly smaller than the later great kiva, which is much more obvious and has a diameter of about 18 meters.  Una Vida is a very confusing and poorly understood site, so the lack of clarity regarding its great kiva(s) is not really surprising.

The only other early great house, in addition to these three, at Chaco is Peñasco Blanco.  It apparently has four great kivas, none of which has been dated.  It’s quite possible that one or both of the two great kivas in the plaza dates to the 900s, but neither has been excavated.  It is also possible that there are additional early great kivas either underlying the later ones or elsewhere in the site.  The number of apparent great kivas is one of the many reasons I think this site is likely much more important to Chaco than is usually appreciated.  It is both one of the earliest sites at Chaco and one of the largest, and it may have served as an important connection to the communities downstream on the Chaco River, where many of the early great houses were, as well as with the Chuska Mountains beyond.  Van Dyke has little to say about it in this chapter, which is understandable since the great kivas are unexcavated (as is the rest of the site).

Snow at Kiva A, Pueblo Bonito

The upshot of all this is that there probably was at least one great kiva built at Chaco in the 900s, and there may have been more, but it does seem to be true that great kiva construction increased dramatically after around 1030.  This is the same time that a lot of other changes were happening in the canyon, including massive construction projects of various sorts at several great houses, and it is probably the time when Chaco first became the regional center for the San Juan Basin (though it had likely been an important center for a long time).   Van Dyke argues that part of this was the appropriation of the great kiva form, which in previous times had been particularly common in communities to the south, by emergent local elites attempting to legitimate their increasingly hierarchical authority and control over periodic regional gatherings in the canyon that were beginning to draw pilgrims from throughout the Basin (and perhaps beyond).  In another article she argues that this process was part of a “tipping point” or “qualitative social transformation” that changed a predominantly egalitarian society into a more hierarchical one.  In this context, the use of great kivas may have been an attempt to establish links with the past by incorporating an old, traditional architectural form into the new and potentially threatening form represented by the great house.  I’m not sure I buy this entire story, but I think at least parts of it are likely true and it’s certainly thought-provoking.

Great Kiva at Lowry Pueblo, Colorado

Wherever they came from and whenever they became part of the Chacoan architectural repertoire, by the height of the Chacoan era great kivas were among the most standardized parts of the highly standardized Chacoan “system,” whatever it was.  There are plenty of puzzles remaining about them, as is true with most everything associated with Chaco, but regardless of whether we are ever able to answer all the questions they pose they are still among the most impressive achievements of this very impressive society.
Van Dyke, R. (2008). Temporal Scale and Qualitative Social Transformation at Chaco Canyon Cambridge Archaeological Journal, 18 (01) DOI: 10.1017/S0959774308000073

Read Full Post »

Kivas in East Plaza, Pueblo Bonito

As I mentioned in the previous post, one of the ongoing debates in Chacoan architectural studies concerns the function of the round rooms that are very noticeable and numerous at the excavated great houses in the canyon.  The standard interpretation for many years, which is still fairly common among archaeologists and nearly universal among the general public, is that these are “kivas,” a type of structure that is assumed to be primarily ritual or ceremonial in function and analogous to similar structures at modern pueblos.  Some archaeologists, however, led by Steve Lekson, have argued that the small round rooms found at virtually all archaeological sites in the northern Southwest up to about AD 1300 are not in fact cermonial “kivas” but rather residential spaces, integral parts of residential “suites” of rooms used for habitation purposes.  The idea here is that in function these round rooms are not the forerunners of kivas at modern pueblos but the descendants of pithouses from earlier periods.  Lekson has been making this case since the 1980s; it was put forth at some length in his Great Pueblo Architecture of Chaco Canyon, New Mexico and reiterated in his recent reworking of one of the chapters in that book for the recently published edited volume The Architecture of Chaco Canyon, New Mexico.  As I will discuss more fully before, other archaeologists seem to increasingly be coming around to this position.

There are a number of other issues surrounding this debate that rarely enter into it, however.  Many of these stem from the rather tenuous and superficial grasp many archaeologists seem to have of the ethnographic and ethnohistoric record.  Basically, the way I see it much of the debate over whether these “round rooms” are “kivas” and whether they were “ceremonial” in function (which are actually separate questions, although they are usually conflated) is based on a whole host of problematic assumptions.

Typical Kiva at Pueblo Bonito

Lekson has been in favor of getting rid of the term “kiva” entirely for small round rooms in the San Juan region before AD 1300 (the qualifications are important, but let’s set them aside for right now).  Instead he calls them simply “round rooms.”  This is somewhat ungainly, but it has the advantage of being unambiguous, accurate, and straightforward.  It also does not presuppose anything about the function of the rooms, which is helpful to tour guides and others who must explain what these structures are without necessarily taking sides on their functions or significance.  On the other hand, however, the term “kiva” is well-established in the literature, and it may be possible to reinterpret it as merely a formal description for a certain type of structure, which would preserve continuity with past scholarship and lead to less confusion for visitors.  Visitors to Chaco when I was working there would often see the round rooms at Pueblo Bonito as ask “Are those kivas?”  This was a tricky question to answer, because in one sense they certainly are what archaeologists have called “kivas,” and it would be confusing to answer “no,” but on the other hand visitors typically have a certain set idea of what a “kiva” is, and answering “yes” tends to reinforce that even if it isn’t necessarily an accurate or useful way to interpret the structures is also problematic.

Kiva at Puerco Pueblo, Petrified Forest National Park

But what is a “kiva,” anyway?  Where does the word come from, and what does it mean?

The short answer is that the term is Hopi, and that its meaning is obscure.  Ki is the Hopi word for “house,” but va cannot be easily interpreted within Hopi (although there have been various attempts to make sense of it), and the word kiva may be a loanword.  Regardless of its ultimate origin, however, in modern Hopi the word kiva refers to a specific type of structure that is subterranean, square (!), and largely, though not exclusively, ceremonial in function.  All the modern pueblos have structures of equivalent function, but they differ in form.  The biggest difference is that the western pueblos (Hopi, Zuni, Acoma, and Laguna) have square or rectangular “kivas,” while the eastern pueblos in the Rio Grande Valley have round ones.  Different pueblos have different terms for these chambers, which is unsurprising because they speak different languages.  It’s interesting that the distinction between square and round forms is geographic rather than linguistic; Laguna and Acoma speak Keres, the same language as the eastern pueblos of Cochiti, Santo Domingo, San Felipe, Santa Ana, and Zia, but they have square kivas like those of Hopi and Zuni whereas the eastern Keres pueblos have round ones like their neighbors who speak various languages of the Tanoan family, unrelated to Keres.  Also, whether round or square, these “kivas” are sometimes, but not always, subterranean.  Among the eastern pueblos, only Taos has subterranean kivas.  The kivas at the other eastern pueblos are entered from the roof, which simulates in some sense a subterranean structure, but they are built above grade.  Similarly, Hopi kivas are generally subterranean or semi-subterranean, but kivas at the other western pueblos are above-ground.  All these structures do have certain formal and functional similarities, however; they all have firepits and entrances by a ladder through the smokehole above the fire, and they were used historically as both ceremonial spaces and something like clubhouses for men.   At most pueblos women are not allowed in the kivas at all, although at Hopi the women’s societies are allowed to conduct certain ceremonies in them.  They are not residential spaces at any modern pueblos.

Sign Describing Kiva at Homol'ovi II, near Winslow, Arizona

When the Spanish first settled in the Rio Grande Valley starting in 1598, they noticed the round chambers in which the local pueblos conducted many of their ceremonies and called them estufas (“stoves”), presumably from the heat of the fires in them.  This term persisted as a generic term for centuries, and the early Anglo archaeologists who began exploring the ancient ruins in the area in the late nineteenth century generally called the round, subterranean chambers they found “estufas”; this is the term George Pepper used in his descriptions of the excavations at Pueblo Bonito in the 1890s.  Use of the term “kiva” rather than “estufa” is probably due largely to the influence of Jesse Walter Fewkes, a towering figure in early Southwestern archaeology and ethnography, who did extensive ethnographic fieldwork among the Hopis as well as excavations of prehistoric sites in the Hopi country and at Mesa Verde.  In one important article on the kivas at Spruce Tree House, Mesa Verde, Fewkes even said:

The special chamber set apart by Pueblo Indians for ceremonial purposes was called by the early Spanish discoverers an estufa, or stove, a name no doubt suggested by the great heat of the room when occupied. An estufa is commonly designated by the Hopi Indians a kiva, which term is rapidly replacing the older name. It is found that prehistoric ruins as well as modern pueblos have kivas and that specialized rooms of this kind likewise exist in cliff-dwellings.

There is a lot of other interesting stuff in this article on kiva form at Mesa Verde, which is useful as a comparison for kivas at Chaco.  In this context, however, the most important thing is just to note that the archaeological assumption that kivas were ceremonial is already here in 1908, as is the belief that the small round rooms of the San Juan region are equivalent in function to modern Hopi kivas despite their different shapes, sizes, and features.  In the course of justifying his argument that the vertical shafts associated with these chambers were for ventilation, Fewkes says:

Modern Hopi kivas, which like those of Spruce-tree House are subterranean, but unlike them in being quadrilateral, have no air  vent except the kiva hatchway. As a rule quadrilateral kivas are much larger and their roofs higher than those of circular kivas, so that the ventilation, which is also facilitated by a more capacious hatchway, is not a matter of great concern.

Note that the assumption that the “kivas” at Spruce Tree House are identical in function with those at Hopi even though they are round rather than square, small rather than large, and equipped with ventilation shafts.  This assumption became virtually universal among archaeologists working in the Mesa Verde and San Juan Basin regions over the course of the twentieth century, until Lekson came along to challenge it in the 1980s.

Kiva at Spruce Tree House, Mesa Verde

When Lekson did begin to challenge this conventional wisdom, he did so on several grounds.  For one thing, whereas a modern pueblo typically has between one and six kivas for the entire community, and this ratio holds true for most late prehistoric sites as well, pre-1300 sites in the San Juan region tend to have many more.  Indeed, as early as Fewkes’s day T. Mitchell Prudden was exploring southwestern Colorado and finding numerous small, modular sites with a few rectangular rooms, a kiva, and a trash mound, which he proposed were basic units of architecture in the region.  To this day these small sites are often called “Prudden units,” and it is generally accepted that even larger sites in the region are composed of large numbers of them aggregated together.  And yet, for many decades the assumption was that these small kivas may have been associated with individual households or extended families but that they were still identical in function with modern kivas, which serve whole villages or large subsets of them, and the additional simplifying assumption crept in that those functions were more or less exclusively ceremonial.

Obviously, however, these “kivas” had developed out of the earlier pithouses, which were clearly residential structures.  So the story came to be that in Basketmaker times everyone lived in pithouses, sometimes with adjacent surface storage rooms, then at some point around AD 800 or so came the “pithouse to pueblo transition” when everyone suddenly moved out of the pithouses and into larger versions of the storage rooms, instantaneously converting the pithouses to exclusively ceremonial use.  Then nothing happened for a thousand years, and when Anglo anthropologists showed up in the nineteenth century they encountered a  simple, communal way of life that had been unchanged for centuries.

Basketmaker Pithouse, Mesa Verde

I exaggerate, but only a little.  What Lekson pointed out was basically that this story sounds nice but doesn’t really match either the archaeological or the ethnographic evidence.  There was a huge series of migrations and cultural changes around AD 1300 that involved the abandonment of the Mesa Verde and Chaco regions and a massive influx of people to the Rio Grande Valley as well as to the Hopi and Zuni areas, and the end result was a settlement pattern across the region that had few similarities to that obtaining in the Four Corners area in Chacoan times (ca. 1030 to 1130) or shortly thereafter.  From 1300 on things look very similar to what the Spanish encountered in the 1500s, but rather different from what they looked like earlier.  Suddenly there were only a few kivas in each village, and they were big.  Not as big as the earlier “great kivas,” generally, but certainly bigger than a typical kiva at Pueblo Bonito or Spruce Tree House.  Also, especially in the west, many of them were square or rectangular.  This probably indicates influence from the south, where square “great kivas” had long been present in settlements of the Mogollon culture.

Kiva at Homol'ovi II, near Winslow, Arizona

Extrapolating backward from this change, Lekson argues that the small “kivas” so common at Chacoan and Mesa Verdean sites were not the forerunners of the later kivas in the large, aggregated pueblos the Spanish found but rather the last of the pithouses, which had a long history as habitation structures in the region and were present at earlier sites in similar numbers.  One implication of this is that the number of round rooms at a given site may be an index not of the importance of that site as a “ceremonial center” but of the resident population.  The idea is not that these round rooms were the only space used for residence but that they were part of “suites” of square and round rooms that served as units of household residential space, basically the “Prudden unit” concept but without any assumptions about which rooms were used for which specific household functions.

This argument has been gaining increasing acceptance in recent years, especially now that careful excavations of many sites in the region are coming up with clear indications of residential use of “kivas.”  Some of these excavations have demonstrated that there really wasn’t an instantaneous “pithouse-to-pueblo transition” but rather that pithouses, or structures with features intermediate between Basketmaker pithouses and “pueblo” kivas, continued to be used well into the eleventh and twelfth centuries.  At site 5MT10010 on Cowboy Wash in southwestern Colorado, for instance, notorious for the claims of cannibalism there, the excavators reported that:

[t]he presence of domestic artifacts, storage features, and mealing bins indicate residential use of the pithouses. In contrast, the surface structures contained few interior features; hearths were small, informal constructions; and no mealing bins were identified. Consequently, the surface structures were probably used seasonally for a narrower range of activities than the residential pithouses.

This site dates to around AD 1150, shortly after the decline of Chaco.  The elaborate masonry round rooms at Chaco great houses were certainly fancier than the humble earthen pithouses occupied by the unfortunate inhabitants of 5MT10010, but there is no reason to assume that they had radically different functions.

Kiva J, Pueblo Bonito

I think Lekson’s arguments are very convincing.  I’m less sure about whether it makes more sense to rebrand these rooms as “round rooms” rather than “kivas” (as Lekson does) or to redefine “kiva” as a formal rather than a functional designation (as many other archaeologists who agree with Lekson’s basic argument seem to be doing).  One possible way out might be to use the term for “kiva” from one of the other pueblo languages, perhaps one of the languages spoken in the eastern pueblos that have round kivas more similar to the Chaco and Mesa Verde ones, but it is quite hard to find information on those languages, partly because their speakers are very wary of outsiders studying any aspect of their culture including language.

There’s much more to say about kivas (or whatever we want to call them) at Chaco specifically, but this post is long enough.  The point I want to make here is really just that interpreting these structures and understanding what they are is a much thornier problem than it seems at first glance.
Billman, B., Lambert, P., & Leonard, B. (2000). Cannibalism, Warfare, and Drought in the Mesa Verde Region during the Twelfth Century A.D. American Antiquity, 65 (1) DOI: 10.2307/2694812

Fewkes, J. (1908). Ventilators in Ceremonial Rooms of Pre Historic Cliff-Dwellings American Anthropologist, 10 (3), 387-398 DOI: 10.1525/aa.1908.10.3.02a00020

Read Full Post »

Old Bonito and Pueblo del Arroyo through the Rubble of Threatening Rock

As I’m sure regular readers have noticed, I haven’t been posting much here over the past few weeks.  As is usually the case, this is because school got in the way; this semester was particularly busy for me because it was my last one, and I had little time for blogging.  I graduated about a week ago, though, so I now have plenty of time to read and blog.  And with the job market the way it is, I’ll probably have quite a bit of time on my hands for at least the next few weeks.

To ease myself back into thinking about Chaco after this recent hiatus I’ve been reading The Architecture of Chaco Canyon, New Mexico, edited by Steve Lekson, which is part of the series of publications forming the capstone to the Chaco Project.  It’s a bit uneven, as edited volumes tend to be, but some of the chapters are very good summaries of the state of knowledge of certain aspects of Chacoan architecture as of 2007 (when the book was published).  It’s definitely not an entry-level sort of book, and I would only recommend it to people who have already read some of the introductory literature on Chaco, but if you have the requisite background it’s a very good way to get up to speed on architecture issues.

Old Bonito from West Plaza Showing Plaza Kivas in Foreground

Among the more interesting chapters are Lekson’s on great house form (an update of a chapter from his seminal book on the topic, which is still an essential reference for Chaco architecture), Tom Windes‘s on early great houses in the San Juan Basin, and Ruth Van Dyke‘s on great kivasJill Neitzel‘s chapter on the history of architectural research at Pueblo Bonito is a bit weaker; it’s fine as a summary, and makes some good points, but doesn’t really go into much depth on the more interesting questions.  The chapter on Chetro Ketl, by Lekson, Windes, and Patricia Fournier, is more interesting, not least because of the obvious tension between Lekson and Windes on certain points of interpretation and how that gets reflected in the text.  I’m still reading the book, so I can’t say much about the remaining chapters, but they seem to be more interpretive than descriptive.  The one I’m currently reading is by Wendy Ashmore and promises to be something of an outsider’s take on Pueblo Bonito, from a scholar who has specialized in Maya architecture and planning (and who has some ideas in that area that have been criticized by others, most notably Mike Smith).  The next chapter, by John Stein, Rich Friedman, Taft Blackhorse, and Rich Loose, is probably the most infamous and controversial in the book, from what I’ve heard, so it should be interesting to see what it says exactly.  The last chapter is a reprint of a paper by Anna Sofaer that was printed in an earlier edited volume, and I find it particularly interesting that Lekson (presumably) decided to give her the last word, given how controversial some of her theories are among Chaco archaeologists.

I’ll have some more thoughts on the specific chapters in the book in subsequent posts, but for now I just want to mention a few key issues that the chapters I’ve read so far have brought up.  One is the recurrent issue of exactly what the function of great houses and their constituent parts was.  Over the past few years there has been an ongoing debate over whether they were elite residences or ceremonial centers (the older assumption that they were aggregated communities analogous to modern pueblos only being held by a few diehards like Gwinn Vivian), but as Neitzel points out for Pueblo Bonito it’s become increasingly clear that they were both, and the real question, to the extent that there is a real question underlying this debate, is which function was dominant, where (both within a given great house and among different ones), and when.  One of the major points Neitzel makes is that this debate is both tedious and increasingly unmoored from reality, and that the whole typological approach is probably best avoided.  Questions of “is it or isn’t it?” are not very helpful at this point, whether applied to a single great house or to the Chacoan phenomenon as a whole, and the use of vague typological categories like “ceremonial center” only exacerbates the problem.

Rooms around Kiva K Partly Overlying Kiva L, Pueblo Bonito

One aspect of the functional question that does seem to finally be getting the attention it deserves is the temporal dimension.  It seems pretty clear that at least some great houses changed in function over time, though the extent to which this was the case in general is unclear (as, indeed, is the question of whether “great house” is actually a meaningful category of structure).  Windes makes clear in his chapter that as far as we can tell early great houses outside of Chaco were primarily residential in function, although few have been excavated so it’s hard to be certain about this.  The data for the earliest great houses inside the canyon is not much better, even for the ones that have been excavated, but it also seems that they were probably initially residences.  There is also substantial evidence that many great houses had residential functions at the end of their occupation, in the twelfth and thirteenth centuries.  The big question is the extent to which great houses were residences during the “Chacoan era” of roughly AD 1030 to 1130, when the Chaco system reached its height and new great houses were being built all over the place.  These buildings probably had multiple functions, and it seems plausible that at least some had residential functions, but it’s frustratingly difficult to tease out the uses of specific rooms and spaces during this period, partly because neither built-in features (often dating earlier) nor artifacts (often dating later) are necessarily related to uses at this time.

A related issue is the function of kivas (or “round rooms”) specifically.  Lekson sees them as primarily residential, with the ceremonial functions traditionally attributed to them on the basis of (rather lazy) ethnographic analogy being a later development.  I find his arguments pretty convincing, and they seem to be increasingly accepted among other archaeologists, especially to the north in the Mesa Verde region, where the sheer numbers of kivas associated with all sorts of sites, and the careful recent excavation data from many sites, is making it increasingly clear that small kivas were indeed residential structures.  This attitude is also beginning to be adopted further south at Chaco itself, and many people on the interpretive staff at Chaco have been incorporating Lekson’s arguments into their tours, which is important because visitors very often ask about this.  I, of course, talked up Lekson’s interpretation a lot when I gave tours at Chaco.  There are some issues about the functions of great houses that arise from this interpretation, however, and the chapters I’ve read so far in this book really only allude to them.  I’ll talk more about this later.

Casa Rinconada, Looking West

Part of Lekson’s reinterpretation of kiva function is the idea that “great” kivas are fundamentally different from small round rooms, and really did have the ceremonial or community-wide integrative functions traditionally implied by the term “kiva” (there some issues with terminology surrounding these interpretations that I won’t go into now).  Certainly the uniformity of the Chacoan great kiva form, and the marked differences between great kivas and small kivas, is consistent with this argument, and Ruth Van Dyke’s chapter is a good collection of the information available about great kivas in the canyon.  Only a few of these are exacavated, and not all of them have been well-documented, which makes interpretation a bit hazardous.  The general uniformity of Chacoan great kivas both in and out of the canyon is striking, however, and it argues for a considerable amount of standardization and a key role in the Chacoan system.  The relationship between great houses and great kivas is another perennial topic of debate, partly due to the limited dataset available.  Van Dyke argues that the association between the two forms is relatively late, and that the building of great kivas at great houses such as Pueblo Bonito starting around 1040 was an attempt to justify new societal arrangements associated with the rise of the Chaco regional system by associating them with the much older great kiva form.  There may well be something to this, but there is also some evidence for great kivas associated with the early great houses at Chaco in the 900s, which complicates the argument.  I’ll do a post on this later.

Finally, one issue that arises particularly in Windes’s chapter on origins is regional variations in architecture and the possible implications of this for migration and connections to areas outside of the central San Juan Basin.   There has been an increasing consensus that there is some connection between the collapse of the Pueblo I villages in southwestern Colorado in the 800s and the huge increase in population in the San Juan Basin to the south around the same time, and similarities between sites like McPhee Pueblo in the Dolores area and early great houses like Pueblo Bonito seem to imply a cultural connection of some sort.  Connections to the south have been studied less, but Windes points out that many early great-house communities have sightlines to prominent southern landmarks such as Hosta Butte, rather than northern ones like the Huerfano.  As far as I know no one has done a really systematic study of which architectural forms correspond to which regions, and the ability to do such a study would be complicated by the numerous migrations over time that are becoming increasingly apparent in the early archaeological record of the region.  It would be a useful thing to try, however.  Again, I’ll do a more in-depth post on this later.

Clearly there’s a lot to say about Chaco architecture and community planning, and with my newly acquired graduate degree in planning I suppose I am well-situated to address these issues, which I intend to do in the coming days and weeks.  However, there are also other topics I want to cover in future posts, including coal, cotton, and chocolate.  Stay tuned.

Wijiji Trail, Heading Back toward Campground

Read Full Post »

Older Posts »


Get every new post delivered to your Inbox.

Join 79 other followers