SCIENCE IN ARCHAEOLOGY

The real purpose of scientific method is to make sure Nature hasn't misled you into thinking you know something you don't actually know.

Robert Pirsig

 

The year was 9268 B.C. The man was trying to make a Clovis spearpoint. He crouched on a flat spot at the foot of the chalk cliffs, near Rattlesnake Buttes in northeastern Colorado. "Coughing twice, rubbing his fingertips on his chest, he lifted the heavy rock and studied it for the last time." His tools had been purposefully selected. The hammerstone was ovoid and of a grainy texture: "It was the possession he prized most in his life." Carefully flaking through the cortex, he struck flake after flake, until the core was exhausted. "He dropped his hammerstone, threw back his head and winked at his helper: 'Good, eh."' Then he reached for his boomerang-shaped soft hammer, fashioned of antler. Gradually, the amorphous flake was "roughed out" into the unmistakable shape of a spearpoint. Taking the elk antler, he pressure-flaked the edges, fashioning "a scimitar-sharp edge around the entire point." Finally, he used his chest-crutch to force off a flake running half the length of the point. The fluting complete, the craftsman stood back and admired his finished Clovis point.

Once the point had been mounted on a sturdy wooden shaft, the craftsman summoned his friends to ambush a mammoth at the neighborhood waterhole. "The mammoth took one faltering step and dropped dead. Not once in a hundred times could a hunter reach a vital point with his spear; usually death was a long-drawn process of jabs in the side and chasing and bleeding, requiring two or three days. But this was a lucky blow, and the men howled with delight." (From Thomas 1979).

"SCIENCE IS SCIENCING" (Leslie White 1949)

That was how novelist James Michener described incidents in the life of Clovis man in his best-seller Centennial (1974). Using the medium of his craft language-Michener created a slice of American life 1 1,000 years ago. The Clovis spearpoint, to Michener, is not an inanimate artifact made of gray-brown stone. The Clovis point is "the finest work of art ever produced in the Centennial region ... a prime fact of our intellectual history." The Clovis craftsman was not a brow-ridged, apelike knuckle dragger; he was "indistinguishable from other men who would occupy this land ten thousand years later ... he had considerable powers of thought, could plan ahead, could devise tactics for hunting ... he did not take himself too seriously." This description is imaginative to be sure, but probably not too far from the mark. While not facts, Michener's inferences are credible, given what we now understand about the archaeology of the prehistoric Clovis culture. As an artist, James Michener grappled with a segment of human experience and attempted to render it intelligible to others. Through his art, Michener attempts to assist modern man in adjusting himself to his environment by understanding his human past.

Like James Michener, artists are characterized by what they do: they attempt to confront human reality, then explain it to others by illustration. Generalities are illustrated by specifics, universals by particulars. Michener wanted to explain what it was like to be a Clovis hunter; he did so by considering one particular hunter in 9268 B.C. at Rattlesnake Buttes, Colorado.

As strange as it might seem, modern scientists share Michener's ultimate objective, to render the human experience intelligible. But scientists approach this goal from the opposite direction. To a scientist, particulars are important only for what they can tell about the universal. In a scientific mode, Michener's mammoth, the Clovis point, and the prehistoric flintknapper are relevant only for what they can tell us about the Clovis culture in general.

So, science is an attempt to abstract universals from particulars.

So is religion, but science is committed to explaining things in naturalistic terms without recourse to supernatural beings and forces (except computers).

THE SCIENTIST IS ONE WHO USES THE SCIENTIFIC METHOD TO ABSTRACT UNIVERSALS FROM A WORLD OF PARTICULARS - IT'S THAT SIMPLE.

These universals are called LAWS

Not like legal laws that are open to interpretation by Johnny Cochran or Janet Reno, but are inviolable - THEY ARE STATEMENTS OF WHAT ACTUALLY HAPPENS, THROUGH TESTING THAT WITHSTANDS OBJECTIVE ANALYSIS.

They are better called Lawlike Generalization, because future investigation can modify earlier generalizations. We imperfectly understand the world around us. The public has a hard time understanding this.

They not only describe the present, but can predict the future. , i.e. Newton's law of gravity.

Archaeology predicts past events based on generalizations from past knowledge.

And although all sciences endeavor to generate their own laws, some disciplines have a much more difficult time, particularly those dealing with human behavior.

Two important attempts to generate lawlike generalizations in anthropology generate great debate still: theory of natural selection and theories of the progressive evolution of culture.

This is mainly because human culture does not necessarily need to follow the laws of nature, and our preoccupation with war and genocide are good examples.

BUT, that does not mean that we cannot endeavor to observe the past using a scientific method. Unfortunately, science in archaeology is often better done than discussed. Today, archaeologists who are ardently anti-science practice the methods of science in their research and zeal to find non-scientific answers.

THE SCIENTIFIC METHOD IN ARCHAEOLOGY

Scotty MacNeish's use of the method.

I. Establish a Hypothesis

a statement that goes beyond mere description of known facts

-through the process of induction, where the conclusion contains more information than the premise - from the particular to the general

 

hypotheses can be derived from any source (a bad dream, hallucination, ravings of a post-modernist, ravings of an empircist). In archaeology, using analogy from other projects, ethnology, history, is usually helpful.

 

What's important is how well the hypothesis accounts for unobserved phenomenon.

In many case multiple working hypotheses will apply to the same data, and help insure a good fit.

II. Determine the test implications

In the classical hypothetic-deductive procedure one used deductive reasoning to seek the logical outcomes.

A deductive argument is where the conclusions must be true given that the premises are true. If a…then b… This is the opposite of induction. Here one predicts a generalized outcome from a specific observation.

Bridging the gap between these statements is crucial. The linking argument in the hard sciences derive directly from known properties. In archaeology we often must use prior knowledge or analogy. In MacNeishes case he knew that corn cobs would preserve for millennia in dry caves rather than open sites. This was an implicit bridging assumption that archaeologists use all the time.

Archaeologists lack of ability to generate adequate bridging arguments is due to our lack of explicit theory of human behavior.

III. Test these implications by further observation

Finally, testing the hypotheses by further observation can take many forms (field and lab work are the observational testing we do in archaeology). Here we can use scientific technology based on general laws and so be more comfortable with the testing.

SW Archaeological Obsidian Project

 

The Scientific Process

SOME ASPECTS OF PROBABILITY AND SAMPLING

Probability and sampling is at the very heart of effective generation of testable hypotheses.

I'm not going to explore this is great detail here (Dr. Hall's class is doing that now), but I do want you to understand some important basic concepts in sampling and probability that will help you understand the reading and the class overall.

It is completely misunderstood by most archaeologists and graduate students:

statements like: 5% is a good sample level in archaeology

there is no use sampling because the materials archaeologists work with are not representative anyway.

These are both false statements

Sampling is the selection of a number of elements from a larger population of elements (sometimes called a universe) for purposes of making certain kinds of inferences about the population overall.

Why sample?

1. too expensive to analyze the entire population

2. we don't want to destroy the entire site

3. we can often determine more about a population from a sample than from studying the entire population, because we can study the sample with greater precision and care than the whole population.

A sampling of definitions:

Simple random sample: selection such that each unit in the population has an equal probability of selection

Stratified random sample: selection of separate simple random samples from sub-populations (the strata can be judgementally of probabilistically derived) - sampling areas of a site

Judgemental or purposive sample: a sample based on judgment, usually based on prior knowledge - early selection of obsidian specimens for chemical analysis based on megascopic criteria.

Haphazard or grab sample: selection based on no statistical criteria

Probabilistic sample: a sample based on statistical criteria

Representativeness: Very important. Only random sampling can be used to yield representative samples.