A shard of pottery from Pueblo Bonito did not smell like chocolate. No brown drink waited at the bottom of a jar for a camera. The important trace had entered the clay long ago and had to be sought by chemical analysis. In 2009, archaeologist Patricia Crown and chemist W. Jeffrey Hurst reported a marker of cacao in selected fragments from Chaco Canyon. The result opened a question about beverages, exchange, and ceremony in the ancient Southwest. It did not reveal a recipe, a merchant’s route, or the name of the person who used a particular vessel.
The National Park Service’s account of the work gives the study a vivid arc: unusual tall vessels had intrigued researchers for more than a century; a small sample from later excavation debris was tested; three likely cylinder-jar fragments yielded theobromine, a chemical marker associated with cacao. The original research paper explains the analysis and its limits. Read together, they show how an archaeological discovery can be exciting without being a tale of someone pulling a perfect treasure from a hole.
Chaco’s cacao is a fitting end to this series because the discovery depended on several kinds of care. The pottery fragments had to retain enough context and material for later questions. A researcher had to form a testable hypothesis. A laboratory had to detect a trace not visible to the eye. A park and museums had to conserve collections. Public interpreters then had to decide how to tell the result without turning an entire ancestral society into a single surprising beverage. The most memorable headline is only one step in that longer chain.
An unusual shape in an extraordinary place
Pueblo Bonito is one of the great houses of Chaco Canyon in northwestern New Mexico. Its rooms, plazas, masonry, and long building history have drawn generations of study. The NPS overview describes a regional network of related communities, great houses, and roads rather than an isolated building in the desert. Interpretations of the network’s political, economic, and ceremonial roles continue to change. A claim about one class of vessels belongs in that wider world, not in a story of a solitary exotic object.
Cylinder jars stood out even among Pueblo Bonito’s varied ceramics. Their tall, narrow form was unusual in the Southwest. Late nineteenth-century excavators found a remarkable concentration of them in one room; the NPS account gives 111 jars from that cache. Fewer than 200 examples were known across the region in the park’s summary. Scholars had proposed different uses, including storing special items or serving a ritual purpose. The shape invited comparison with vessels elsewhere, but resemblance was not proof of what the jars held.
Most of the famous complete jars went to the American Museum of Natural History in New York. That movement is part of the modern history of Chaco: major early excavations removed enormous quantities of material, while collections and records were divided among distant institutions. The NPS collection history explains that the park itself does not hold the principal early Pueblo Bonito collections made by the American Museum and later National Geographic expedition. A researcher interested in one great house may have to follow its materials across several repositories.
The jars’ old interpretations were not foolish guesses to be swept away by chemistry. Each represented an attempt to explain an uncommon shape, restricted distribution, and archaeological association. New evidence can make some explanations more likely and others less. It should not turn a provisional result into an all-purpose answer for every vessel. A jar might have had more than one use in its lifetime, and not every tall jar necessarily contained the same drink. Archaeology improves by specifying what was actually tested.
Five pieces, three signals
The core of the 2009 study was small and precise. Crown selected five pottery fragments for organic-residue analysis. Three were characterized as probable cylinder jars; the other two had different or less certain forms. Hurst used analytical chemistry to look for absorbed compounds. The published paper reports theobromine in the three likely cylinder-jar sherds and not in the other two. The researchers connected that signal to cacao use. This is the finding, stated at the scale of the actual sample.
The NPS page says the pieces dated approximately A.D. 1000–1125 on stylistic grounds. That is an interpretation of when the pottery was made or used, not a stopwatch on the moment a drink was poured. Pottery style, deposit context, and chemical trace each answer a different question. The power of the study came from bringing them together. Its limit comes from the same specificity: five pieces cannot directly tell us how common cacao consumption was across Chaco, who drank it, or whether a single ceremony followed a fixed script.
The molecule was found in the ceramic fabric, not as a visible puddle. This matters because an absorbed residue can outlast the obvious contents of a vessel. It also means the analysis required careful attention to contamination and comparison. The paper explains its laboratory method and why theobromine was treated as a marker. A reader does not need to operate the instrument to follow the reasoning: identify a substance associated with cacao, test relevant samples, compare positive and negative results, and make a claim no larger than the evidence warrants.
The result was a surprise within a known network of long-distance connections. Other materials, including copper bells and scarlet macaws, had already shown links between Chaco and more southerly regions. The cacao plant grows in climates far from the high desert, so its presence raised new questions about how materials and ideas moved. A chemical trace does not identify a single caravan or prove direct travel by a Chaco resident to a particular Mesoamerican city. Exchange can pass through many hands and places. The “route” is a research question, not a line that appears in the sherd.
The people outside the laboratory
A headline such as “Chocolate discovered at Chaco” is effective because modern readers know chocolate. It can also carry the wrong picture. The ancient beverage was not a present-day candy bar, and the people who used it were not waiting for a modern scientist to give their life meaning. The paper uses cacao because it names the plant and material implicated by the residue. A public article can explain the familiar connection while refusing to project today’s sweet product, packaging, or habits backward by a thousand years.
The discovery story should also include the people whose ancestors made and used the vessels. Chaco Culture National Historical Park says that Pueblo communities, Hopi, and Navajo clans have ancestral or traditional relationships to Chaco, expressed through oral histories and kinship. The exact nature of each connection differs. A laboratory paper about theobromine does not replace those histories, and an outside writer should not invent a ceremonial interpretation on behalf of living communities. It can say what the chemistry shows and point readers toward public descendant perspectives.
The jars themselves were once part of decisions made by ancient potters and users: how tall to shape a vessel, what to paint, where to keep it, whether to use it again, and where it finally rested. Those decisions were made within a community, not inside a future museum. A fragment detached from its field record may still contain a residue, but it loses some of the spatial and temporal setting that helps interpret the result. The ability to test archived sherds is therefore an argument for careful collection records and against looting. Both the material and its documented origin mattered.
Modern researchers also work inside institutions with histories. The first Pueblo Bonito excavations took place under standards different from current archaeology. Objects traveled to New York and Washington, and the park’s own collection is incomplete for that early work. Later researchers had to use old field documentation and undertake new, limited sampling. A neat discovery headline can obscure this uneven archive. It should not. The path by which evidence survived influences what questions can be asked today.
An old trench and a new question
The NPS account explains that a University of New Mexico project in 2004–2007 revisited trenches from Neil Judd’s earlier work in Pueblo Bonito’s middens. From the many sherds recovered, Crown chose five for the cacao test. The new result did not require excavation of a pristine room full of jars. It grew from old research areas, a large body of fragmentary material, and a question that earlier excavators could not test chemically. The story is as much about revisiting an archive as about a moment in the field.
That sequence shows why an ordinary sherd can matter. An intact decorated jar is photogenic and may carry more market value. A broken fragment with a useful record can answer a question about use that an unprovenanced showpiece cannot. If researchers know where a fragment came from and how it relates to a vessel type and period, laboratory analysis can give it another layer of meaning. Without that information, a positive chemical test might prove only that an unknown pot touched cacao at some unknown time.
New techniques are often described as magic keys to the past. They are tools with costs and limits. Residue analysis consumes a small amount of sample. Instruments require controls, and interpretation requires comparison with known materials. Other methods such as remote sensing can reveal patterns without digging but still depend on the questions asked and on what the ground allows a sensor to register. The NPS page mentions such technologies while noting that the park rarely sponsors large excavations today. Restraint and innovation can go together.
The decision to sample also has a history. A curator must know whether a sherd is stable enough, what material will be removed, how the test will be documented, and where its results will be kept. A scientist must decide which comparison pieces will help distinguish a signal from chance or contamination. The institutions and communities connected to Chaco may have interests in the proposed work and in how its results are presented. None of this is as eye-catching as a positive theobromine result. It is the framework that makes the result credible and gives others a way to evaluate it. The laboratory finding is an event, but the preparation and recordkeeping make it knowledge.
The research also demonstrates that old collections must be cared for, not merely stored. A sherd whose label is lost can become almost mute about its place. A collection whose objects are inaccessible cannot easily be reexamined. Museums and parks need conservation, inventories, archives, and policies for study. They also need consultation with descendant communities about research and display. Keeping a fragment available for science is one responsibility; deciding whether a proposed analysis or publication is appropriate may require more than the researcher’s enthusiasm.
From evidence to inference
The paper’s strongest claim is chemical evidence of cacao use in Chacoan ceramics of the period studied. A more detailed story—that a particular group drank cacao in a specific ritual, imported it by a certain route, or restricted it to elites—requires additional evidence. The distinctive jars and limited distribution make some interpretations plausible. The NPS explanation presents ritual and restricted use as possibilities, not a witnessed event. That distinction can be kept clear without draining the result of interest.
Mesoamerican comparison is useful but also has limits. A similar vessel form or beverage practice elsewhere suggests questions to test; it does not establish that Chaco residents copied an exact ceremony. People can adopt, adapt, or transform materials moving across regions. A cacao drink at Pueblo Bonito could have held a local meaning that no surviving object states in words. The jar cannot speak its own social rule. Descendant histories may offer relevant perspectives, but they should be invited and attributed rather than appropriated to fill a gap in the chemical paper.
The five-sample design also warns against an easy generalization. Three positive likely cylinder jars and two negative comparison sherds support a relation between vessel form and cacao in this set. They do not prove that all cylinder jars held cacao or that no other vessel did. Later testing could broaden, qualify, or complicate the pattern. A good discovery story leaves room for those outcomes. It tells readers which result is durable and which exciting extensions remain hypotheses.
Even the word “negative” needs care. The two pieces without detected theobromine are comparison results under a particular method and preservation history. They do not prove no cacao ever touched those vessel types, and they certainly do not describe every kitchen or ceremony in the Chaco world. Samples can be small, residues can decay, and vessels can have multiple uses. The useful contrast is narrower: in this selected set, the probable cylinder jars produced the signal and the other two did not. That pattern justified a new interpretation and a larger research program. It did not transform a small test into a census of ancient beverages.
Nor should a press account confuse “earliest documented evidence known to these researchers at publication” with “first time anyone in a place ever did something.” Archaeological absence is not proof of historical absence. A newly tested object can push back the documented range of a practice without establishing where the practice began. The distinction matters because races to be first invite exaggerated claims and can turn careful scholarship into a contest of headlines.
What the archive cannot return
The cacao study benefited from preservation of pieces and records. It cannot restore every lost association from Chaco’s long collecting history. The NPS collection page notes variable documentation across projects and large holdings outside the park. Some questions are answerable because a sherd’s context survived; others remain uncertain because early work removed or mixed materials. This is an ethical consequence of past practice, not merely an inconvenience for modern research. Every site disturbed without adequate record narrows the future’s choices.
There is also an imbalance in the public story. We can name Crown and Hurst, cite their paper, and describe the laboratory. We cannot name the ancient potter from the tested fragment. The asymmetry should prompt humility, not an invented character. The vessel records skill and use; it does not preserve an individual biography. Descendant communities may hold stories or meanings that are not public or not specific to this jar. A writer can let the unknown remain unknown while still describing a human life behind the material.
Some readers will ask whether the cacao finding proves a “lost civilization” or a secret transcontinental trade system. It does neither. Chaco is a well-documented ancestral landscape with a complex regional system and living connections. Long-distance movement of materials is important without requiring a fantasy of hidden empires. Copper, birds, ceramics, and cacao each have distinct evidentiary trails. Their combination invites questions about exchange and social life; it does not authorize a map of a single road drawn from a chemical result.
The same caution applies to the word treasure. Cacao was valuable, and the jars were unusual, but their value to history is not a price. An old complete cylinder jar removed for sale would have less scientific meaning if its room, layer, associated materials, and collection path were unknown. A modest fragment with a reliable record gave researchers a new answer. It is a reversal of the collector’s logic, and it shows why protecting sites and archives matters more than extracting spectacular objects.
How to tell the ending
The ending is not that scientists have solved Pueblo Bonito. It is that a small test changed one part of the account and created better questions. What other vessels carried cacao? How widely did people use it? Through what chains of exchange did it arrive? What did communities make of it? Which interpretations are appropriate for public telling? Those questions may be answered differently by chemistry, archaeology, and descendant histories. None should pretend to be the only voice.
A responsible public story can be entertaining here. A nearly invisible trace in a broken pot leading to a surprising connection across a continent is a strong narrative. It has a setting, an old mystery, a test, a result, and a changed understanding. It does not need an invented ritual scene, a secret recipe, or a claim that every resident drank chocolate. The drama lies in the gap between an unremarkable sherd and what a careful question revealed.
The six places in this series make the same argument in different ways. At Blackwater Draw, the layer around a point mattered. In the Mimbres valleys, looting destroyed relationships a bowl could not carry away. At Gila, an original beam and wall are best understood in place. At Pecos, descendants directed the return of ancestors taken into collections. At Mesa Verde, associated communities changed how a museum tells their history. At Chaco, an archived fragment yielded new information because enough material and record remained. Archaeology is not only about what people find. It is about what they preserve, whom they include, and how honestly they distinguish evidence from an attractive story.
The Chaco sherd stays small at the end of this account. It does not become a whole banquet, a known merchant, or a complete theory of a great house. It holds a trace of cacao and a connection to a documented place and period. That is enough to be remarkable. The care that kept those facts together is what allowed the discovery to happen long after the pot broke. The next possible discovery may already be in a collection or still in the ground. Its value will depend, again, on whether we leave enough of its context and enough room for living voices to tell us what it means.
Sources and further reading
- Patricia L. Crown and W. Jeffrey Hurst, “Evidence of cacao use in the Prehispanic American Southwest,” PNAS, 2009: original sample, analysis, and conclusions.
- NPS, “Pre-Columbian Chocolate Discovered at Chaco”: accessible account of jars, project history, and interpretation.
- NPS, Chaco archaeological collection: excavation collections and the limits of park holdings.
- NPS, Pueblo period at Chaco: regional system and competing interpretations.
- NPS, Chaco questions and answers: contemporary Pueblo, Hopi, and Navajo connections.
Return to the six-part series hub, revisit the layers at Blackwater Draw, or explore the Treasure Symbols and Maps series for other cases in which a mark means less when separated from its source.