Dr. Leland Gilsen
Organization of Cultures
What is Archaeology?
Sampling & Bias
Presumably you came here to learn about Oregon's past, right? Well, this section offers my in-depth perspective on the prehistory of Oregon and how it fits into this region.
Oregon is where the the NW Coast, Columbia Plateau, Inter-Mountain, Great Basin, and California chaparral ecologies and cultures merge. The people who lived here over the last 14,300+ years have developed technologies to adapt themselves to the demands of these varied ecological zones with their abundant natural resources. Proceed to an in-depth exploration of Oregon Prehistory ».
Value of Archaeology, What Would be Lost?
Why bother with all that digging anyway? An overview of the professional view on why archaeological sites should be protected and preserved. Continue »
Oregon has been inhabited for at least 14,300 years, based on a radiocarbon date on a human caprolite (poop). During this period of time, the climates and ecology of plants and animals have changed. Populations have grown and in turn, the cultures of Oregon's inhabitants have changed. Human land/resource use has changed as adaptive strategies have coped with changing demands. Just about every landform contains some evidence of human use. The relative density of use is reflected in the archaeological record.
Places where human groups performed some activity and left behind physical remains as artifacts or features are archaeological sites. The artifacts and features tell a story about the resources that were valued, methods of resource extraction (productive technology), methods of resource distribution, methods of resource storage (investment), and methods of resource consumption. Faunal and floral remains as well as features give information on the seasons of occupation as well as resource values and changing resource use through time and across space. Archaeological sites are used as a tool to lump activities together into manageable units for analysis. Humans tend to gather together for some activities and to separate for others. People repeat similar actions in specific locations, and the result is a site.
Archaeological sites are fragile because their scientific value is not just in the artifacts and features, but in the patterns of these in relation to each other. The material remains of past patterned activities is the primary data used by archaeologists to reconstruct past human behavior and decisions. Archaeological sites are systemic because they contain data derived form patterned behavior. Any disturbance of the patterns reduces any site's archaeological value. Archaeological sites are non-renewable because the people who created them no are longer alive. The environmental circumstances of the past lifeways are not the same as today. No living culture can truly reflect the past, but they can be used as simplistic models for reconstructing long vanished lifeways.
Did You Know...
Oregon is so complex, that it has five ecozones and related cultural adaptations?
Oregon's past is organized into periods that have a chronological and conceptual framework that are applied statewide with phases embedded in them that have geographical constraints. Human populations appear in the archaeological record at the end of the last Ice Age. The oldest radiocarbon date is the 14,300 year old date from human dung. The Dietz site near Wagontire is a Western Clovis/Western Stemmed occupation represented by fluted and stemmed projectile points. Such sites fall between 12,000 to 7,000 years ago. Sites within this period have been called PaleoIndian, PaleoAmerican, Pre-Archaic, Proto-Archaic, Western Clovis and Western Stemmed. Such site represent economic strategies that exploited (now-extinct) megafauna. Traditional interpretation is towards an unbalanced economic strategy emphasizing fauna over flora. There is increasing evidence for use of ground stone for processing plants during this period, blurring its relation to what some archaeologists call the archaic.
Archaic represents an economic pattern of post-megafauna broad spectrum plant and animal gathering and hunting. This roughly coincides with a shift in climate towards hotter and dryer conditions. Lakes and marshes reduced in size or dried up. In Oregon, the Archaic is usually broken into Early, Middle and Late Archaic adaptations. Early Archaic is equated with a broad spectrum gathering/hunting/ fishing economic transition from megafauna to modern fauna. Plant processing is much more in evidence, suggesting a shift towards more balanced economic systems. The generic period of occupation spans 7,000 - 4,000 years ago.
Middle Archaic is associated with a mature adaptation that includes semi-sedentary occupations sites with houses. The economy is mixed and intensifies over time. Projectile point style changes are associated with this period as well. The collective time period is from 4,000 - the BC/AD interface. This approximately corresponds to a climatic shift towards wetter and cooler conditions as well. Lakes and marshes in the dry areas of the state expanded.
Late Archaic is seen in two ways. It is associated with the introduction of bow and arrow technology or with the development of land management techniques (economic) such as systematic burning, that do not involve direct manipulation of plants or animals. Burning opens up climax vegetation, allowing the natural increase of valued plants and animals as an indirect (and the desired economic) effect. The common time period is from BC/AD to the entry of trade goods from Euro-Americans or actual contact.
At the time of contact between the native cultures of the Willamette Valley of Oregon and the Euro-Americans, most of the drainage systems above the falls at Oregon City were occupied by groups who spoke a common language: Kalapuyan. Chinook was spoken in the Portland Basin and the Clackamas drainage. The Euro-Americans tended to name the river after the groups which lived on that river, for example: the Tualatin, the Yamhill, the Santiam, and the Luckiamute . The words "tribe" and "chief" were used by early explorers, settlers and politicians. "Tribe" has a particular meaning in anthropology. It represents a level of political integration above that of "bands". We do not know the political level of the Kalapuyan groups. One of the aims of archaeology is to reconstruct the social and cultural levels of past groups.
What is archaeology?
Archaeology is the systematic recovery by scientific methods of the material remains of human life, culture and history (or prehistory) from former times. Archaeology is a sub-branch of anthropology, and anthropology is the scientific study of the physical, social and cultural development and behavior of humanity. In other words, anthropologists study people and archaeologists study artifacts and features used by people in the past.
Archaeologists survey the land to locate the artifacts and features left behind by past cultures and societies. Archaeologist record the location of the artifacts and sites. Archaeologists also excavate sites to recover artifacts and to record features. This process is called data recovery. The raw data is then converted to information by many means. Information is interpreted data.
I am an archaeologist by profession. As such, I was trained to write in the American anthropological style. Every profession has its jargon. An automobile mechanic has terms for car parts that are a mystery to me. I have terms that are probably a mystery to that mechanic. I will try to explain terms as they appear in this site. I will try to state complex theories and positions in terms that can be understood by its non-specialist readers.
Archaeological publications use a reference style based on the authors name and date of publications. For example if the following material appears at the end of a sentence "(Allely 1975:7)"... it refers to an article written in 1975 by Steven Allely and the reference is to page 7 in that article. Look at the bibliography for Allely and you will see the reference. If more than one publication is in the same year, the date would be 1975a or 1975b. If more than one author appears, the author listing occurs after single publications by the lead author. This is sometimes referenced by the first author followed by "et al.", meaning "and others.". So the 1987 publication by Minor, Baxter, Beckham and Toepel could appears as: (Minor et al. 1987). Look at a bibliography in one of the papers in the THEORY section to see how this works.
There is jargon in every discipline. I will try to define an anthropological term before it is used for the first time. For example, the numbers "35LK1" or "35-LK-1" or "35 LK 1 or "35LK 00001". What does this mean? It is an archaeological site number. Archaeologists have a long standing habit of naming archaeological sites that have been excavated, but most sites do not have a name. Many states, including Oregon, assign unique identification numbers to all sites. This is done so that artifacts removed from them can be cataloged and traced back to a particular point in space.
This system was originally developed by Nebraska archaeologist named Paul Cooper. He worked during the depression era for the Work Program Authority and developed a county abbreviation for Nebraska to which he added a number suffix. This 2-3 letter county code with number was adopted by the Smithsonian Institution in the 1940's after Cooper joined their staff. The 48 states were alphabetized and numbered and this number was used as a prefix. States began to adopt this system in the 50's and 60's.
Site numbers are read right to left: (Oregon was 35th in that alphabetic system) so 35-LK-1 would be the first site recorded in Lake county in Oregon. Because numbers are hard to remember, and names are easy to recall, most major excavations, as noted, add a site "name," thus 35-LK-1 is "Fort Rock Cave. 35PO23 would be read as the 23rd site recorded in Polk county in Oregon. 45KLT251 would be the 251st site recorded in Klickitat county in Washington."
Oregon County Codes
Many projectile point styles have been named. The term "Clovis" comes from Clovis, New Mexico, where it names both an archaeological site and a fluted projectile point style. Many projectile points are named as it is much easier to remember what an "Elko-Eared" point looks like rather than something like "Point Type 2J."
The term prehistoric simply means before a society has written records. The term "prehistoric" has been misused and often has been stereotyped into an image of the brutal "cave man." In archaeological terms, it simply means before any written records. All humans share a "prehistory." Some Indians dislike the term "prehistoric" because they have oral traditions that extend into the past. When I studied Mesoamerican (Mexico, Guatemala, British Honduras) archaeology years ago, it was prehistory. Now that the glyphs have been translated and it has become historic archaeology.
Sampling creates a bias in our view of human technology. The negative connotation of primitive technology is bogus. Human groups have very sophisticated technologies, but their orientation varies through time. If you learn nothing else from this web site, it should be the concept of "appropriate technology". There are no "primitives", only people like you or me, doing things according to their contexts of other people, places and times. At some time in the future, we could be called "primitives". Are we? No! And the same goes for any time in the past! Forget that concept, it is wrong. Human beings were just as intelligent and observant in the past as they are today and will be tomorrow. How they interpret those observations were, are, and will be, different.
An artifact is some portable object made by human beings: an arrowhead, a hammer stone, a scraper, or a basket are all artifacts. "Features" are non-portable human-made things. A fire-pit, an earth oven, a post hole, rock art or a housepit, are all examples of prehistoric features. Bottles, cans, the flag Betsy Ross made, or a guitar are examples of historic artifacts.
You will see abbreviations in quotes such as "BP", which is another way of saying "before present" or "years ago". Radiocarbon dates are listed as a date with a plus or minus interval. Radiocarbon dates are based on counting the number of decays in carbon atoms then using statistical techniques to average out the results. So a date of 2000 ±50, means the laboratory came up with a date of 2000 years ago and there is a 95% chance that the date lies between 1,950 and 2,050 years ago. They are "absolute" dates, but in a blurred range of probability.
Because the current date is always changing, the date of 1950 (when this dating process started) was picked as the date to add or subtract from the radiocarbon date to get the calendar date. So an uncorrected date of 2000 ±50 radio-carbon years ago is 2000 minus 1950 or 50 BC with a 95% probability of falling within the range from (2050 minus 1950) 100 BC to (1950 minus 1950) zero (the BC/AD change over). A carbon date of 700 would be 1950 minus 700 or AD 1250. You may see 500 ±50 RCYBP which means 500 plus or minus 50 years Radio-Carbon Years Before Present. Since the relative amount of radioactive carbon in our atmosphere has changed over time, there is a curve used to correct dates and the results are called corrected dates. Corrected dates may differ significantly from uncorrected ones. Correction curves have changed over time, including major revisions of older dates in 1999, so I will use raw dates in this web site.
Relative dates happen when it is obvious that one object or feature came before another such object or feature. For example, soil layers accumulate through time and artifacts in lower undisturbed layers were placed there before artifacts in layers above them. Their relative date or position is known, but there absolute date is may not be known. Early Archaic artifacts are consistently found below Middle Archaic artifacts in undisturbed contexts. Diagnostic artifacts are objects with a distinctive form or style that can be relatively or absolutely dated. An example would be car body styles. Many people can recognize the decade of manufacture of a car by a glance. The car body style can be diagnostic for a date range.
There are fundamental sampling issues in archaeology. The older the site, the greater the loss in artifacts due to decay and disturbance. This is simple entropy (increasing disorder or decay in any system) at work. There is what appears to be a growth in the range of artifacts recovered from archaeological sites over time. This appears to show an increasing growth in technology, both in the quantity and quality in artifact forms.... but part of this may be an artificial pattern due to differential decay rates. Older sites contain fewer organic remains from the process of decay over time.
If one takes into consideration the range of artifacts recovered at Ozette (a village covered by a mud slide that sealed the organic artifacts in an oxygen-free grip) then roughly 90% of all of the artifacts recovered by archaeologists in the Northwest (NW) came from Ozette. Based on the Ozette and other wet sites, perhaps only about 3-7% of the artifacts in use in the NW region were made out of stone. Stone artifacts are just about the only things that survive in open sites subject to acid soils and plants that dissolve organic artifacts. One might suspect that the Plateau, Inter-Mountain, Basin and California regions had a greater percentage of stone tools, but even if this approached 10%, it is not much data for reconstruction of human ecology and economics as 90% of the range of organic artifacts and material culture is gone. This makes it difficult to determine the exact material makeup of a group due to differential decay rates between inorganic and organic artifacts and in different environmental conditions.
Wherever past people left behind material traces of their activities, they created an archaeological site. In Oregon, the first archaeologically recorded people were the "PaleoAmericans": the University of Oregon field school conducted excavations at Connley Caves and Paisley 5 Mile Point Caves sites where the oldest directly dated human remains12,400 RCYBP mitochondrial DNA in human coprolitesin the Americas were found. But archaeology is a continuum, that includes both "prehistoric" (before written languages) and "historic" sites and artifacts. Sites are accumulations of artifacts (objects made by people - i.e. arrowheads, dart points, scrapers), features (structures made by people - i.e. hearths, housepits, earth ovens, cache pits), and associated trash deposits (middens - i.e. flaking debris, bones, plant remains, pollen) within a soil matrix. Sites were subject to mixing as they were deposited. They have been subject to disturbance and decay ever since. They are fragile and slowly degrade.
A bias in the archaeological record is changing population size through time. I believe that it is a simple fact that there are more people alive today than in the past. Some anthropologists would say this is a questionable assumption In my opinion this has been a universal trend with minor irregularities due to disease and warfare. There were fewer people around during PaleoAmerican times than in the Early Archaic, so for any set period of time (say 1000 years) there were fewer archaeological sites created by earlier groups than later groups. Thus the sample size decreases as we go deeper into time... there are simply fewer sites to find from earlier periods and thus fewer artifacts and features to recover. As noted above, this means longer periods for entropy (growing disorder) to be at work destroying sites, so fewer intact sites survive with a narrower range of intact or preserved artifacts and features.
This brings up a positive process. Since populations grow, groups must change their adaptations to meet the demands of the population size. I firmly believe that human groups only change in significant ways when they are forced to change and population growth is the pressure to change. The complexity of society thus grows. There are simply more and more alternative specializations available in complex societies. The rate of change in development of specializations also tends to increase over time because populations grow and with larger populations, if nothing else, there is a greater pool for random chance change. But at the same time this also allows one to assume, with greater confidence, that older and older sites spread out across larger and larger space scales and time frames are variations on each other. They did the same things in the same way over larger areas of space and through longer periods of time. This redundancy, or conservativeness in material culture, helps to reduce sampling loss to some extent in older sites.
Beyond the things (artifacts, features, trash), sites contain patterns of artifacts, features and trash. To the archaeologist, it is these patterns, as information, that are of real value, not the "things." When sites are "dug," all artifacts, features and the trash are recorded in three dimensions as raw data. The archaeologists then search for patterns in the recorded data. The patterns reveal how long-vanished, recent, or still-living, populations lived. The "dig" is properly done within a "research design" (a work plan) that details the methods and techniques to be used, and the studies that will be done on the resulting data to produce information.... in other words, how the data will be turned into information.
Archaeologists use two basic rules when digging: 1) use a constant volume unit to measure the content of the site; and 2) keep digging levels until all archaeological data is exhausted. This means dividing the site up into grids and keeping each test unit constant in size (clean straight pit walls) so that the qualities and quantities of data from each unit can be compared to each other. Each square is dug in small levels (or increments), such as 10 centimeters, or in natural layers. A pit is not finished until the excavator reaches bedrock or encounters soils known to precede any human presence. Some archaeologists terminate excavations after two successive levels contain no cultural artifacts, but this can be misleading.
The essence of scientific archaeology is pattern recognition. Experience with the range of literature and training in the study of how living cultures are organized allows the trained professional to see the hidden and subtle patterns not readily visible to the untrained eye. Patterns are subject to decay. There are always natural processes at work that impact such patterns. Also, people reuse places and mix things up. These patterns are then examined within a set of "research questions" generated from the body of literature.
This brings up another fundamental issue. The patterns can be organized by quantity (size) and quality (integrity). Some patterns are small, they are specific to a person or small event. These patterns are rarely preserved in the archaeological record because other nearby events tend to destroy their integrity. For example, the sharpening of a single tool in a camp by knocking off some flakes may be disturbed by other activities by other members of the camp. Broad patterns covering many common events tend to preserve longer and with better integrity, as they are often the sum of many smaller events that are very similar. Each day women may forage for plant foods and return with them to the camp. The accumulation of food remains forms a pattern of economic choices that illustrates what foods were important to this camp at the season of occupation.
In general, the higher the level of integrity of a site and the more sheltered from the elements, the greater is the chance for detailed patterns to have survived. This is why dry rockshelters and caves are so important ... preservation is excellent. Looters know this simple fact, and the act of looting sites where preservation is best is the ultimate act of destruction to the science of archaeology and for understanding our common past. Once a site has been disturbed by looting, the integrity is destroyed and the information value is severely diminished or lost.
The Indians of Oregon made the majority of their stone tools out of rock that chips with a conchoidal fracture (like the breakage pattern in glass). Obsidian is volcanic glass. Glassy basalt was used as well as materials popularly known as "gemstones": jasper, quartz, or agate (these are all cryptocrystalline silicates - meaning they have a minute crystal structure and are therefore glass-like). Because rocks do not decay very fast, the flaking debris from tool manufacture is a hallmark of most prehistoric sites. Often lost or discarded broken tools are also found such as arrowheads, dart points, scrapers, and utilized flakes. Sometimes ground stone artifacts are found such as manos, metates, pestles, mauls and mortars. Most tool-stone is brought to the site, so "foreign" stone (manuports) may mean an archaeological site is present.
The Willamette Valley Indians also heated rounded river cobbles in earth ovens and then placed the hot rocks into skin or animal stomach bags to boil foods. They may have placed the hot rocks onto leaves to steam foods in a pit. The rock would shatter, creating what is called "fire-cracked-rock" (FCR). FCR is usually found in clusters and has many irregular angular edges and is often discolored.
The plant and animal remains were discarded at the site as well. Charred organic remains (plants, animal remains) can be preserved for thousands of years in non-acid soils. Pollen from plants brought in by to a campsite lasts in soil for thousands of years as well.
Quarries, where the stone was mined, are also found. Quarries associated with some cryptocrystalline silicates show extensive heat treatment to improve the quality of the raw material. The raw material was placed in shallow pits and roasted. This changes colors and makes the material waxy looking. Potlid fractures (small craters) are a hallmark of heat treatment, and is probably the source for the popular myth that Indians flaked hot stone with drops of water. It is easy to look at potlid fractures and assume water was dripped on the stone to produce the flake. In reality, they are caused by water flashing to steam under the surface of the material as it is heated.
Other typical sites types include rockshelters and caves; peeled trees where the bark was stripped for sugar or fiber; pithouse villages with shallow circular depressions where the houses once stood; rock art; stone alignments including cairns; fishing weirs; and burials.
Where people made pottery from fired clay, ceramics become an important part of the material record. There is a low-key ceramic tradition along the souther Oregon coast, as well as some ceramics in late Great Basin cultures in Oregon.
Not every site contains sufficient data to study the basic who, what, when, where, how and why questions of basic reporting. Often, only parts of the archaeological record are present. Some sites are a part of an overall economic system. Luckily, people repeat actions at many locations so there is a bit of built-in redundancy. Many sites at many locations that were occupied at roughly the same time must be studied to reconstruct a lifeway. Archaeological research is additive and new theories and models are generated as data is gleaned through research. Some questions cannot be dealt with until other questions are solved. Slowly, through planned research, models are created and tested to explain the data. In all cases, the intact prehistoric archaeological site is the key for unlocking the past.
One way to visualize the problem of prehistory in Oregon is to imagine each 100 years as a jigsaw puzzle for each of about 100 different native groups or bands. Each puzzle piece represents an archaeological site where different families within that group did slightly different things in different places at different times. The last 10,000 years would be 100 such puzzles. Then imagine that all of the puzzles are mixed together and many of the pieces are hidden, lost, torn or partial. At contact, there were at least 50 different language groupings. That means there are 5,000-10,000 puzzles, each overlapping, all mixed up, and with missing, lost, torn, and hidden pieces. This is a simplistic model of the problems that archaeologists face in Oregon.
An archaeologist must first find a piece (the site), figure out which puzzle it may belong to (cultural affiliation). They then try to figure out where it goes into the: who, what, when, where, how and why mentioned above. But in reality, each piece is in bad shape (is if each puzzle piece was broken, torn, had faded colors and edges that no longer fit tight). It may be that the problem has no complete solution, so every site takes on importance.
But there are ways to fit this mess together to create a coherent whole. Because there are common sense aspects to where people settled (flat ground, sheltered areas, near water and wood for cooking, and near valued minerals, plants and animals), archaeologists can make assumptions about the poor data using models from the still-living descendants of the people who lived in the Willamette Valley. Ecotones, where different ecozones merge have a greater variety of plant and animal life, and are places people used. Dry meadows were burned to expand and maintain valued plants. In general, the Indians were minimizing risks and maximizing gains in relationship to their physical, biotic and cultural environments of cooperating and non-cooperating other human beings. Sites are found just outside the flood zone along river bottoms or flood free tributaries. Fishing sites are found near chutes and falls where anadromous fish concentrated. Quarries are found near outcrops of obsidian, glassy basalt and cryptocrystalline silicates. Aboriginal trails followed burned ridge tops where travel was easier and camps are found on the saddles. Areas near seeps and springs are likely to have sites. Major river terraces outside the normal winter floodplain were used for winter villages. Stable wet meadows in forested areas produce valued foods not tied up in tree trunks as found in climax forests.
Any resource used by a society has certain characteristics: relative abundance; relative predictability; and relative mobility. Minerals are static, except where rivers carry cobbles down stream, but their relative abundance and quality (another measure of predictability) varies from deposit to deposit. A campsite near a quarry with high quality obsidian may be far from local food supplies.
Plants tend to be abundant, predictable and non-mobile. Animals tend to be rarer, less predictable and much more mobile. Most plants produce seasonal edible parts, and some are more abundant and predictable than others (camas, cattail, acorns, hazelnuts, berries, bitterroot, etc). Some animals are plant-like (shellfish) and anadromous fish runs can be very abundant, fairly predictable and be constrained in mobility by river chutes and falls.
A good economic strategy is a mix of plants and animals using gathering and hunting techniques the minimize risk and maximize gains. Most prehistoric groups had a seasonal round that exploited many different areas, plants and animals through the year. Burning was used to modify plant and animal regimes. Places where important minerals, plants and animals occur are common sense places where sites are found. But there are always those who do illogical or non-strategic things. Archaeology relies on sampling and probabilities to arrive at conclusions. Complex systems are full of feedback loops that make linear outcomes less likely, and make our conclusions slippery and vague. This is the nature of archaeology as a discipline. Given enough time and research, the science will oscillate and converge on a "probability model" that will reflect some level of reality. Just where we are on that convergence, only time will tell.
Harold Mackey's book on the Kalapuya incorporates "the findings of amateur archaeologists." One of these is the report by a farmer named John Berg who supposedly unearthed an irregular circle of large phallic stones on the brow of a hill south of Salem. Mackey noted that professor of geology D. C. Livingston had indicated that the monoliths had been shaped by nature but then modified by "the crude tools of some prehistoric race" (Mackey 1974:49). The stones were reputed to be evidence for a crude form of phallic worship arranged into a temple-like form. While we were discussing a possible collaboration to revise his book, Dr. Mackey and I visited the place where the stone monoliths had been discovered in the hills south of Salem. The hill was surrounded by hexagonal basalt columns, but in a relatively impure from. The columns appear to be in a circle around the hill, but this is an erosional process. When the soil stabilizes below the tops of each column, there is a freeze/thaw operation in the soil that splits off fragments of the relatively impure hexagonal columns that gives an appearance of a phallus. It is my opinion that this is entirely a natural phenomenon, and it continues to be in operation today. Several such new phallic-like stones had been placed near the entrance road to the farm by the owner. New ones are being created to replace those that break off and fall down the hillside. There was no evidence that the columns are anything except a natural erosional process in an unusual geological situation. There was no temple-circle of phallic stones constructed by the Kalapuya.
There are, however, phallic pestles found throughout the Willamette Valley (see Plate 15, page 65 in Mackey's book for a slightly phallic pestle). In addition, the stone mortar bowls can contain repeating V-shaped vagina-like symbols around their perimeter (see Plates 10-12, pages 59-61 in Mackey's book). The processing of foods in the bowl may have been a symbolic fertility-increase ceremony ensuring that valued plants will remain available through time. There is the possibility that the appearance of such symbolism is relatively late, and also represents the increased competition for the remaining wild foods in the valley.
The amateur lore that the Kalapuya were mound builders or built "burial mounds" also needs to be addressed. While there are sites along the Calapooia and Muddy drainages that appear to be mound-like, survey along any of the rivers and streams indicates that winter camps were placed on slight natural rises in the ground to escape the winter floods. With no water control structures in the mountains (dams), the river systems would flood each year. One can still look around the landscape during floods to find slight rises that remain above mean high water. The Indians camped on these locations over and over, bringing back to the camps plant and animal foods as well as wood for fires, stone for ovens, and wood and bark for house construction. Gradually, over time, these places built up layers of organic remains and discarded tools, and flakes.
In one way, the mounds are like tells (a hillock formed by the accumulation of debris, earth, or other material on the site of an ancient settlement). The sites were higher and dryer than the surrounding ground so they were also often used as winter village locations. A longer stay during the winter means that the random chance of a death was a bit higher and thus burials are more likely in the "mounds". Land away from the site was water saturated, so burials tended to be within the site/tell itself, giving the appearance of burial mounds. There is no evidence the mounds were constructed, rather they just grew out of a long slow process. There is evidence that several such selected locations were centered around camas patches and were used over and over. As one site became too polluted with human feces and/or overrun by fleas, the alternate locations were used. Rainfall, decay of the feces and the cold of winter killing the fleas would gradually clean up the old locations so they could be used again. Early explorers noted this process in winter villages along the Columbia. So these were not burial mounds, they were locations where people lived, and their "mounding" resulted from concentrated human activities over long periods of time.
Throughout this site will be many quotations. Some of the early writers used terms that are now seen as derogatory. One of these is the use of the word squaw. When quoting, I will use the original wording, but this does not mean I approve or condone the use of terminology that modern ethnic groups feel should no longer be used. There are also spelling issues with early writings. I will use the spelling as in the originals.