In 1994, in a field in southwestern Quebec, archaeologists confirmed that they were standing on top of a village. Not a camp, not a scatter of tools, but a village with houses, with rubbish heaps, with the ground still holding the outlines of where people had lived, farmed, argued, cooked, and buried their broken things. It dates to the middle of the fifteenth century, which means it was already old ground before any European had set eyes on that part of the continent. The Canadian government has described it as the largest pre-contact indigenous village site ever found in Quebec.

They excavated a small portion of it, and from that small portion came roughly 150,000 artifacts. When researchers took that density and applied it to the parts of the site they had not touched, the estimate they arrived at for what is still down there was more than seven million. Seven million objects sitting in the ground in a field right now, almost none of which is going to be dug up. That last sentence is the part that surprises people, because the obvious assumption is that this is a funding problem or a scheduling problem.

It is not. The decision to leave most of that site in the ground is deliberate. It is considered, and once you understand the reasoning, it is very hard to argue with. There is a second thing worth saying at the start, because it changes how you hear the rest of this.

Nothing about that field looks remarkable. There is no mound, no ruin, no standing stone, nothing that would make you slow the car down. Everything that makes it one of the most significant archaeological sites in the province is below the surface of a working landscape. Invisible.

And it was invisible to everyone for roughly five centuries. But first, that number needs explaining, because seven million sounds like a made-up figure until you know what an artifact actually is. When most people hear the word artifact, they picture a museum object, a carved pipe, a complete pot in a display case with a small card next to it. That is not what is being counted here.

In archaeology, an artifact is any object recovered that was made, modified, or used by people. A single broken cooking pot does not count as one artifact. It counts as every fragment of that pot, and a large ceramic vessel that has been shattered and trodden into the ground over centuries can produce hundreds of individual pieces. The same applies to stone.

Making a stone tool is a subtractive process. You start with a lump of raw material and strike flakes off it until you have the edge you want. Every one of those flakes is an artifact. A single afternoon of somebody sitting down and making one tool can leave behind a scatter of dozens or hundreds of pieces of debris.

Add animal bone, broken up during butchery and cooking and then discarded. Add charred plant remains. Add fragments of clay pipes, beads, worked bone, and the countless small pieces of everything that gets used and dropped and stepped on. Now put several hundred people on that spot, living there for years, doing all of that every single day.

Seven million stops sounding absurd very quickly. It starts sounding like arithmetic. So what was actually there? The people who built this village were St.

Lawrence Iroquoians. They were Iroquoian-speaking, agricultural, and they lived along the St. Lawrence River Valley in villages, not in small mobile groups. They are related to, but distinct from, the Haudenosaunee nations to the south and the Wendat to the west.

Their economy was built on farming, and specifically on corn, beans, and squash grown together. A combination that works because the plants support each other. The beans return nitrogen to the soil, and together they provide a diet that can support a large settled population. Farming at that level is what makes a village of that size possible in the first place.

At this site, archaeologists identified 25 structures and two middens. Those structures were longhouses, and they are one of the most impressive pieces of vernacular architecture in North America. A longhouse is a long narrow building made of a bent wooden frame covered in bark sheets, not a hut. Some ran to tens of meters in length.

Inside, a central corridor ran the length of the building with a line of hearths down it, and families lived in compartments along both sides, typically with two families sharing each fire. That arrangement is not just a building. It is a social structure made physical. Who you live next to, who you share a fire with, which house you belong to, all of it laid out in the architecture.

Longhouses were organized around related women and their families in societies where lineage and property passed through the mother line. It is worth trying to picture the inside of one, because the numbers alone do not convey it. You walk in through an opening at the end. The building is dim, lit mostly by the line of fires running down the center and by the smoke holes above them.

It smells of wood smoke, because everything does, permanently. Along both walls are raised platforms used for sleeping and sitting, with storage above and below them, and food and equipment hanging from the frame overhead. Corn is stored in quantity, because the whole year depends on it. And it is full of people.

Multiple families, several generations, children, in a single continuous space with no internal walls in the sense we would recognize, only the divisions of who belongs where. Then step outside, and there are 24 more of those arranged in a settlement with fields cleared around it. Twenty-five of those in one place is a substantial settlement. And then there are the two middens, which sound like the least interesting thing on the list and are actually the best.

A midden is a rubbish heap. For an archaeologist, that is close to a jackpot, because everything a community throws away is a direct record of what it was doing. Broken pottery tells you about cooking and storage and style and trade. Animal bone tells you what was hunted and fished and when.

Charred seeds and corn kernels tell you what was grown and eaten. Failed and broken tools tell you what was being made. Better still, middens build up in layers. Material thrown out later sits on top of material thrown out earlier.

So a deep midden is not just a pile of objects. It is a sequence, and it can record how a community changed across the decades it lived there. There is one more thing middens do that is easy to miss. Because they accumulate continuously, they capture the ordinary rather than the exceptional.

Most of what survives from the past is skewed toward the valuable and the ceremonial, because those things were made carefully, kept, and deliberately placed. A rubbish heap has none of that bias. It is a record of the unglamorous majority of life. What people actually ate on a normal day.

Which pots broke. Which tools failed. What was so common that nobody thought about it. That is precisely the information that written history is worst at preserving in any culture, anywhere, and it is why a deep midden can be more informative about how a community lived than a spectacular single find.

Which brings us to a genuine mystery, and it is one of the strangest in Canadian history. In the 1530s and 40s, Jacques Cartier traveled up the St. Lawrence and encountered these communities directly. He visited a village at the site of what is now Quebec City and another further upriver at what is now Montreal.

He described the second one as a substantial fortified town with longhouses inside a palisade and cultivated fields around it. Thousands of people, organized, settled, farming. Then, roughly 60 or 70 years later, Samuel de Champlain came up the same river. They were gone.

Not reduced, not relocated a short distance. The St. Lawrence Iroquoian villages that Cartier had described were simply not there anymore, and the valley had been taken over by other groups. What happened in that gap is still debated, and it is worth being careful here, because this is a subject where confident storytelling has caused real harm.

Several factors are proposed, and they are not mutually exclusive. Warfare with neighboring nations in a period when competition over trade routes was intensifying, as European goods began moving inland ahead of Europeans themselves. Epidemic disease, which moved through the continent along trade networks far faster than colonists did. Environmental and agricultural pressure.

And dispersal, meaning that communities broke up and their members moved into and joined neighboring nations. That last one matters, and it is the point most often lost. The disappearance of the St. Lawrence Iroquoians from the archaeological record of that valley is not the same thing as the disappearance of those people.

Current understanding leans heavily toward dispersal and absorption, meaning their descendants are present today within Haudenosaunee and Wendat communities. A settlement pattern ended. A people did not. So there is a village sitting in a field in Quebec that was occupied before all of that happened, holding an estimated seven million objects from a society whose villages were emptied within living memory of first contact.

And we are not digging it up. Here is why, and this is the part that reframes the entire subject. Excavation is destruction. That sounds like a provocation, and it is simply a description of the method.

An archaeological site is not valuable because objects are in it. It is valuable because objects are in it in a particular arrangement. What is next to what? What is above what?

Which post hole lines up with which? How deep something sits? That spatial information is the actual data, and the objects on their own are a fraction of it. The moment you excavate, that arrangement is gone.

You can record it meticulously with measurements and photographs and drawings, and good excavation does exactly that. But you cannot put it back. You get one attempt, and after that the ground is empty, and all that survives is your record of it plus a great many boxes. Which means every excavation is a bet that your methods are good enough to justify permanently consuming the evidence.

And the history of archaeology is full of sites where that bet lost badly. Consider what has become readable in the last few decades. Residue analysis can now identify what was actually cooked in a pot by analyzing the chemistry absorbed into the ceramic. Isotope analysis of bone can tell you what an animal or a person ate and where they grew up.

Ancient DNA can be recovered from bone, from teeth, and increasingly from sediment itself. Microscopic wear analysis can tell you what a stone tool was used to cut. Radiocarbon dating has improved enormously in precision and in resistance to contamination. None of that existed when many major sites were excavated in the twentieth century.

Material was dug out, washed, cataloged, and stored, and in the process a great deal of information that we could now read was scrubbed off or discarded as irrelevant, because nobody knew it was information. So the modern position is straightforward. Dig the minimum necessary to understand the site and to answer specific questions. Leave the rest where it is.

Not because it is unimportant, but because it is important enough to be worth waiting for. The ground is a better archive than a warehouse. In situ, undisturbed, the material keeps its context and it keeps its chemistry. In a box, it has already lost most of what made it worth having.

There is a second reason, and in Canada it now carries as much weight as the first. Decisions about indigenous heritage sites are not made by archaeologists alone. They are made with the descendant communities whose ancestors are in that ground, and those communities have a direct say in what happens, whether excavation proceeds, and how any human remains and belongings are treated. That is a substantial change from how this work was done for most of the last century, and it is the reason a lot of material stays where it is.

And there is a practical reality on top of both. One hundred and fifty thousand artifacts from a small portion already represents an enormous processing burden. Every object has to be cleaned, cataloged, analyzed, stored, and conserved permanently. Excavation is the cheap part.

The rest lasts forever and costs money forever. Digging seven million objects out of the ground would not be a triumph. It would be an institutional crisis. Now, the honest limits.

The seven million figure is an estimate, not a count. It comes from taking the artifact density in the excavated portion and extrapolating across the rest of the site. That is a reasonable method, and it is how these projections are done, but density is not uniform across a village. Some areas will be far richer than others, particularly the middens, and some will be nearly empty.

The true number could be substantially different in either direction. The site is also not sealed away in perfect safety. Buried archaeology is not immune to damage. Plowing disturbs upper layers.

Drainage changes and shifting water tables alter the chemistry of the soil and can accelerate the decay of organic material. Freeze and thaw cycles move things. Looting is a permanent risk at any known site. Leaving material in the ground protects its context, but it does not stop time.

And the word lost, which appears in a lot of coverage of sites like this, is doing something unhelpful. The village was not lost. It stopped being occupied. The location dropped out of the written record kept by the people who arrived later, and it was subsequently identified by archaeologists.

Those are different things, and the framing of a lost civilization waiting to be discovered belongs to adventure stories rather than to a real place with living descendants. But the core of it holds, and it is a genuinely strange thing to hold in your head. There is a field in Quebec. If you drove past it, you would see a field.

Under it are the outlines of 25 longhouses, two rubbish heaps holding the layered record of decades of daily life, and an estimated seven million objects made and used and dropped by people who lived there before Cartier came up the river. A small fraction of it has been examined. The rest has been deliberately left alone by people who understand exactly what is down there and have decided that the best thing they can do with it for now is nothing.

Not because it does not matter, but because it will still be there when we are better at reading it.