Developing New Interpretations from Old Data at Montezuma Castle National Monument

View from Montezuma Castle, MONTEZUMA CASTLE NATIONAL MONUMENT.  Red cliffs sweep from the right to the left distance, with green mountains in the background and a blue green river to the left.
View from Montezuma Castle,

NPS photo.

Matt Guebard, NPS

October 15th, 2015

The Castle A site located within the Montezuma Castle NM was initially excavated in 1934 by NPS archeologists Martin Jackson and Sallie Pierce. They concluded a catastrophic fire destroyed the site long after abandonment, an interpretation that has persisted for over 80 years. A reanalysis of ceramic data coupled with archaeomagnetic dating suggests this initial interpretation is incorrect.

Open Transcript


Karen: Our second speaker, Matt Guebard, is going to talk to us about a reanalysis of an excavation of site in Montezuma Castle National Monument. For over 80 years, the archeological community has accepted the excavators’ interpretation of the Castle A site, which was that the buildings were destroyed by a fire long after abandonment. Matt has a different take on the data. I'm going to stop talking so that he can tell us what he's learned. Thank you for being here with us today, Matt.

Matt: Thank you. Thanks everybody for sticking around. We had some technical difficulties earlier, so Karen is going to be helping me advance the slides today.

I'll be presenting the preliminary results of an ongoing study at the Castle A archeological site which is located in Montezuma Castle National Monument. As you'll see, the reanalysis of old data and the collection of new data has led to an interpretation that is almost entirely different than the one that was used by the National Park Service for over 80 years.

Montezuma Castle National Monument was created in 1906 to protect and preserve the Montezuma Castle Cliff Dwelling. It's one of the first groupings of national monuments that was created by the President following the passage of the Antiquities Act. The monument is located in the town of Camp Verde which is right in the geographic center of Arizona. You can see it here on this map.

This elevation depicts the Montezuma Castle cliff face. It includes the iconic Montezuma Castle Cliff Dwelling which is on the right, which is to the east in the real world, and then Castle A which is to the far left.

We're going to be talking mostly about Castle A today. Just to orient you a little bit more, here's a photograph of what Castle A looks like along with a planimetric map. I show this just to illustrate how complex the site is. There's multiple levels, 6 stories in fact, and really a mixture of standing masonry architecture, rooms built in alcoves and cavates which are human-created cave spaces.

This is an artist's reconstruction of the site. The artist who drew this is actually the famous New Mexico Zuni archeologist and anthropologist, Edmond Ladd, who worked at the park in the 1960s. I like this drawing because, again, it shows the multiple stories and the complexity of the site, but it also shows that the east and west walls and the primary roof beams are tied into or abutted directly to the cliff face. This means that the north interior wall of each of those rooms is actually the cliff itself. It's pretty ingenious and a complicated blueprint.

Most of what we know about the site comes from Civil Works Administration excavations that were conducted between December of 1933 and April of 1934. They were headed up by a young U of A masters student named Earl Jackson, who went on to have a Park Service career. He also hired another young student named Sallie Pierce. You may have seen her in the literature as Sallie Pierce Van Valkenburgh or Sallie Pierce Harris. She also spent her entire career in the Park Service. They excavated the site with the help of 10 what they called unskilled laborers. Those were probably local, out-of-work folks. As you can tell from the photograph, they're using pickaxes and flat shovels. It's a pretty rudimentary excavation technique that they referred to as “room clearing” or “room cleaning.” That's where they were just removing architectural debris and a lot of cultural deposits from within the room spaces.

They excavated a total of 9 rooms and several test trenches throughout the site. The intent of the project was really to uncover standing walls for stabilization and eventual interpretation, as well as to recover what they were calling museum quality artifacts, the type of stuff that they could display in the visitors center.

One of the things that they found was burned roof material in every room without evidence of previous looting activity. That's 7 of the 9 rooms excavated. This led them to conclude that the site had been destroyed in a large, what they called a conflagrative fire, that caused it to actually detach from the surrounding cliff face and collapse. Remember I mentioned that it was attached into the cliff itself, so the idea was that the fire was so large and destructive that it caused the pueblo to actually peel away from the cliff face and then fall onto itself.

Karen: Can I interrupt you?

Matt: Yeah.

Karen: I left this slide up so we might be one slide out of sequence, but what is flat surface that we're looking at on the left side of this slide? Is that a terrace floor? Or is it a rooftop?

Matt: It's actually a reconstructed roof. In a lot of our sites, in the Southwest particularly in our unit, one of the things interpretatively was to reconstruct rooms and put roofs on them and then have a living history kind of thing going on. That roof was built after the excavation just to give folks a sense for what it might have looked like in one of the room spaces.

Karen: I see. Thank you.

Matt: Jackson and Pierce felt that the weight of evidence, in this case, 2 to 4 inches of lime dirt, that was right on top of the occupational floors but below the burned roof material, proved that the fire had occurred long after the abandonment of the site. Basically, they were basing their entire interpretation of site abandonment on 2 to 4 inches of lime dirt on only a few floors.

I spent a considerable amount of time looking through the archives to try to find additional photographs, like the one shown here, that might illustrate the extent of that 2 to 4 inch accumulation. In looking through the old excavation photographs, I also began to see what might be called de facto refuse. That's usable artifacts left in their original use locations, and often right on top of the floor surfaces. There's a number of reasons why you might see de facto refuse in a room. One common explanation includes large or unexpected fires where inhabitants wouldn't have the time to collect their belongings before they can escape the site.

With this in mind, I then went back through the field notes to try to find any kind of reliable descriptions of artifacts on floor surfaces or trash deposits in the rooms. The idea was to try to reconstruct a history of room use. What I found was that a lot of the descriptors for stratigraphic deposits, of which there were several, really weren't standardized and they were different depending on what version of the report you read or if you were looking at the field notes. I did find some descriptions of artifacts on floors in 3 rooms and indications of a trash deposit in at least 1 room.

Looking back through all of this existing information, I think there are 3 basic points to make. First is that, based on inconsistencies in information pertaining to the stratigraphy that was originally recorded at the site, Jackson and Pierce's descriptions of stratigraphic deposits really aren't a reliable indicator of when the site burned. Their post-abandonment fire hypothesis is questionable. Secondly, trash deposits indicate a complicated history where people are using, abandoning, and reusing rooms for a long period of time. This is something that we would expect to see in a large pueblo with a large occupational history. Again, their interpretation of abandonment based on stratigraphy is probably a little bit overly simplistic. Lastly, de facto artifacts strongly suggest that the site may have been occupied at the time of the fire.

With all that in mind, we started looking actually on site for evidence of the large fire that was described by Jackson in the excavation report. Originally, what I hoped to find was a location where archeomagnetic or thermal luminescence might provide a basis for either confirming or refuting the post-abandonment fire hypothesis. What we ended up finding was this location here, outlined in yellow. You can see the discoloration. This is in room 2 which is at the far east end of the site.

If we zoom in a little bit closer there, you'll notice that not only is the bedrock and the building stones discolored, but the surrounding mortar is that orangey-red color. It's got that nice black and gray rind on it, indicating that it's been oxidized. This room has been heated to a pretty high temperature.

In July of 2011 and June of 2013, Tom Windes, who is the guy in this photo here, he's my partner on this project, and I collected some archeomagnetic sample cubes from the oxidized mortar in room 2. Tom and I collected 20 one-centimeter cubes, that's a 10 cube sample set for each year that we collected. Then we sent them to the New Mexico Office of Archaeological Studies' Archaeomagnetic Lab.

We received the results back in December of 2011 and then November of 2013. Reported dates represent comparisons between each specific sample year and 2 virtual geomagnetic poles or VGP curves, including the SWCB2000 and the Wolfman curves. Date ranges associated with each curve represented possible ranges for the fire in room 2. As you can see, there were several dates returned for the 2011 samples, including those in the 10th, 11th, 14th, 15th, and 17th centuries. Then, the 2013 samples returned only one set of date ranges. I think it's important to note that the 2013 samples are considered to be more accurate than the 2011, although both are very accurate. Additionally, the Wolfman curve is also often considered more robust from about A.D. 1000 to 1400. This suggests that the Wolfman date ranges for 1375 to 1415 and 1370 to 1395 are the most accurate representations of possible dates for the fire at Castle A.

So,14th century archeomag dates match really surprisingly well with late-dated, decorated ceramics originally excavated from the site in 1933. What we were able to do was take the latest dated ceramics found at the site and crossmatch them with our archeomagnetic dates. Taken as whole, the ceramic assemblage represents a period of time between about 1315 and 1450. Overlapping dates for ceramics and archeomagnetic dates encompass a slightly smaller period between about A.D.1375 and 1395.

So far we've been able to build a pretty convincing case that the site burned between 1375 and 1395 when it was occupied. We also began looking at other lines of evidence reported by Jackson and Pierce, but subsequently overlooked in their final report. One example was the discovery of an articulated human skeleton in room 3A. The skeleton was found on top of the floor surface but below the burned roof material, in exactly the same location as the 2 to 4 inches of lime dirt that provided the basis for their interpretation of site abandonment. It's also interesting to note that the room also had a de facto artifact deposit, including basketry, tools, and ground stone. This, again, provides further evidence that the room was occupied at the time of the fire.

Interestingly, Jackson also noted evidence for violence in the form of a cranial vault fractures on two individuals interred in one single grave just outside the Castle A site boundary. The fractures were consistent with what might be expected for traumatic injuries associated with violence. The human remains also displayed evidence of having been burned. It's interesting, I think, to consider the fact that they're buried in the same grave. This could suggest that they both died in the same violent event and that it's associated with a fire as well.

Following the preliminary results from the project, we contacted the Monument's culturally associated tribes and we invited them to participate in the interpretation of the project data. We sent out a blanket request to all of our tribes and we received responses from two groups that represent three separate cultural groups that include the Hopi, the Yavapai, and the Apache. Each group was invited to participate in separate on-site consultation meetings in 2013. Then we followed that up with a group meeting in 2014. Before each meeting, the tribal members filled out standard release and consent forms to be video and audio recorded. Initially, we asked them to provide information about the site, including traditional site names and then any related oral histories that they might have.

Interestingly, each group had a separate story explaining the events that we were seeing in the archeological records. For instance, the Hopi have stories relating the experiences of groups that were living at the site before and during a violent attack where fire was actually used as a weapon. Somewhat similarly, the Apache and the Yavapai related stories of a alliance wherein ancestors from both of those groups attacked the residents of the site and then used arson to evict them. Although each story also contained elements that weren't evident in the records, things like the participation of supernatural beings or use of magic, the core of the story, I think, provides some important information about social conflict and intentionality behind evidence for violent behavior. Most notably, the story suggests that disputes with cliff dwellers, which were probably ancestral Pueblo people, likely over land ownership, led to a violent conflict and then the forced eviction of those people from the Verde Valley.

My hope is that this project will contribute to a more complete archeological understanding of the Verde Valley in the 14th and 15th centuries. Considerable debate has centered around evidence for pre-contact violence throughout the Verde Valley and the result of this project may contribute to that discussion.

We're also in the process of studying the construction history of Montezuma Castle Cliff Dwelling as well as partnering with the nearby Forest Service sites, they also have cliff dwellings, and trying to place our sites and the events that we're seeing in the record into a larger context. There's more to come on that.

Similarly, oral histories from Apache, Yavapai, and Hopi participants suggest that ancestors of these modern groups were interacting with each other during the 14th and 15th centuries. This is an idea that archeologists dismissed over 50 years ago. My hope is that this project will help to restart discussions about the cultural diversity of the area during the pre-contact period and help to open a dialogue with other archeologists and tribal members about what ancestral populations might have looked like in the Verde Valley, and of course, also, the usefulness of oral histories in helping to interpret the past.

This project highlights the importance of reanalysis at publicly accessible sites, but it also shows the intersection of tribal stories and archeology. One of the big questions here is how does the National Park Service interpret this new information without focusing too much on violence. As you guys all know, we have what amounts to interpretive curriculums which often include a variety of information about things like what did people eat? What kind of clothing did they wear? How did they build their houses? How do we make sure that one dramatic story like this one doesn't overshadow all the other ones that we want to tell our visitors.

Also, similarly, a lot of the interpreters that I've spoken to recently have told me that they worry that this new interpretation might bolster stereotypes about Native Americans or pre-contact folks in general. This is probably a really legitimate concern. To address that, we're working with the tribes and really trying to determine how to most appropriately tell the story without focusing too much on violence or conflict. Ultimately, this is their story. Everything considered, the consulting tribes want us to talk about this story and they want to highlight the intersection between archeology and their stories, but they also want to vet all of the information. The conclusion here is that we still have a lot of work to do before any of this information makes it to thewayside signs. I'm pretty optimistic that this project will result in information that can be interpreted in an appropriate and useful way.

There's been a lot of people that have been involved in this project from the start. This is only a handful of folks that have been involved. They've provided feedback and assistance with field work and sharing information. Thanks to all those folks for being involved and thanks to everybody out there for listening in.

Any questions?

Speaker 3: Hey, Matt, this is Anne. Did you find that there were stories or components of stories that the tribal representatives couldn't tell you or didn't want to share with you because they hold some significance or that there's only certain individuals who are basically allowed to have that information?

Matt: Yeah, most definitely. I was told that we got the very short, watered down version of the story and that there was probably a lot more to it, but that wasn't going to be shared with us. What we got from all the participants was really just enough information to confirm what we were seeing with the archeology. I'm sure there's a lot of other details that are involved that we just don't have access to. Which is okay.

Speaker 3: Yeah, I guess that's how I interpreted that. It's like, that's okay, I don't really need to know it all, I just need to know that it's important. Whatever they feel comfortable telling me. That's interesting.

Matt: Yeah.

Karen: This is Karen. It's really interesting that you thought they were fighting over land. What made you think that?

Matt: That's what I was told. That's what a couple of the tribal representatives told me.

Karen: Hmm. Have you tried to relate this to climate change at that time period at all? I don't know my Southwest archeology. Is this around the time of those really killer droughts?

Matt: There are some things going on climatically, but I think in our area, particularly being right next to Beaver Creek and the Verde River, I'm not sure that these groups would have been impacted. There's a lot of migration stories, from the Hopi particularly, that talk about groups moving around on the landscape, not always related to environment, often related to a number of other reasons that are maybe more social than environmental.

My understanding is that this period of time, at least archeologically, there's not ... the population is not so high in the Verde Valley. There are groups that are on their way out. There's a lot of movement around the valley. Interestingly, the stories from some of the tribal members said that folks living in cliff dwellings, as opposed to other open air sites, were some of the only folks from this group, from the Ancestral Puebloan group, who were left in the valley. Then they were pushed out. After they left, then all of the ancestral puebloans were up north, probably in the locations where they still are today.

Karen: Hmm, how interesting. I wanted to ask also about artifacts on the floors of the rooms. Does this indicate to you that the rooms were not abandoned, as well as the other evidence that you alluses to earlier? The manos and the metates, they would have been such a large amount of effort to create. The fact that they were left there, it speaks to me that they were abandoned during times of stress.

Matt: Yeah. I think that's possible. Some of the other things that were found that are really interesting is, in two of the rooms, they actually found cooking pots that were sitting on the hearths, like they would have had something in them that was cooking. Then groupings of similar tools in corners of rooms, or sleeping mats on the floor in a corner, or basketry in a certain area. There was a lot of ground stone and almost every room had ground stone in it. Then fewer rooms had some of those more perishable ... and items that would be carried off more easily if folks were leaving.

That's just not something that I think that they were looking for in the 30s. They gave it this catch-all interpretation of just ... the fire having occurred long after they abandoned the site based on, again, only the stratigraphy in a few rooms. But what's interesting, actually I should say is that in the report, they do indicate that it's interesting that they're finding these types of artifacts on the floor, but it's really like a sentence in the very end of the report. Maybe they thought about it and for whatever reason they didn't follow up on it. I don't know. Hard to say.

Karen: But the collections were available to you for examination?

Matt: Yeah. Most of them were. Although having said that, the proveniences for a lot of the artifacts were not great. The best we hoped for was at a room level. A lot of times we could match up individual artifacts with the overview photographs from the excavation. The stuff that we were really interested in, particularly the decorated ceramics, there really was very little provenience information for those. A lot of times we knew they came from the site, sometimes we knew they came from a midden as opposed to a room or vice versa, but not the kind of detailed information that we would have wanted to be able to do a really good reanalysis.

Karen: Yeah, but it's a very nice argument for maintaining reasonable museum collections, which is for reanalysis exactly of this type.

Matt: Yeah. I agree. Particularly with the older national monuments, I think that we hang on to the interpretations. There's probably a number of reasons why that happens, but I think there's probably a lot of reanalysis projects that are either going on now or will go on in the future that will really change our interpretations of some of our primary resources. That's pretty exciting to think about.

Karen: Yeah, that is. It's also interesting to think about how you're going to negotiate interpretations of the site in terms of the violence that you think occurred there. I wonder …

Matt: Yeah. Our fear has been that if we interpret it incorrectly, then Montezuma Castle will become ... that's what it will become known for. I've been really reticent in even talking to our interpreters too much about it because I'm afraid it will be sensationalized and the rest of the story will just be lost.

Karen: Yeah. There's something else to think about, too, in terms of the violence in our own society, you know? To balance it against what we see today and provide some sort of a context within to talk about violence. I think the work that's been done in interpreting the internment camps now points to a different direction, a new direction for talking about some of these difficult topics. But maybe not.

Do we have anymore questions for Matt?

Matt, thank you very much for being with us today. Come back next week everybody and we will talk about the Fowey and some of the issues and challenges with investigating and managing underwater archeological resources. See you next week.

Visit our keyboard shortcuts docs for details
29 minutes, 6 seconds

Matt Guebard, 10/15/2015, ArcheoThursday

Montezuma Castle National Monument

Last updated: July 21, 2021