Aristotelian vs. Galilean Forms of Explanation



William S. Abruzzi





"The development of thought since Aristotle could, I think, be summed up by saying that every discipline as long as it used the Aristotelian method of definition has remained arrested in a state of empty verbiage and barren scholasticism and that the degree to which the various sciences have been able to make any progress depended on the degree to which they have been able to get rid of this essentialist method ."


--Karl R . Popper, The Open Society and Its Enemies  1950, p. 206


   Aristotle explained the behavior of an object, such as a rock, in terms of the “essential nature” of that object.  For Aristotle, a non-measurable force existed within an object that compelled it to behave in a certain manner.  A stone, for example, was classified by Aristotle as a heavy object, while fire was defined as a light object.  Since heavy objects, likes stones, tend to fall downwards and light objects, such as fire, tend to move upwards, these behaviors --gravity and levity respectively-- were deemed by Aristotle to be part of the essential nature of those objects. The significant point here is that the factors determining the behavior of an object, according to Aristotle, all originate within the object to be explained, and depend upon the unobservable nature of that object.

   One of the consequences of Aristotle’s method of explanation was that he proposed two kinds of motion: natural motion and unnatural motion.  Stones do not always move downward.  When thrown they may proceed upward before returning to the earth.  This upward motion of the stone was considered unnatural to Aristotle because the force causing the stone to move upwards originated from outside the stone and, hence, did not result from the stone’s own volition (its nature).  One of the important components of an Aristotelian explanation, therefore, is that the behavior which is typically associated with a particular object becomes viewed as the “natural” behavior of that object.  Because rocks tend to move downward, it thus becomes the “nature” of rocks to move downward in Aristotelian philosophy.  We, thus, have two explanations for the movement of the rock:  one explanation for the rock’s downward motion and a very different explanation for the rock’s movement in any other direction but downward.  All of these other movements become “exceptions to the rule.”  The result, however, is that Aristotelians cannot really explain the movement of the rock at all --why it moves faster or slower, at various angles, etc.  Only vague generalizations can be made:  “Rocks will normally move downwards”.










   Then along came what has become known as The Mechanical Revolution.  Initiated by Galileo and later synthesized by Isaac Newton, The Mechanical Revolution resulted not only in a different conception of the physical world, but also in a completely new way of analyzing that world.  It ushered in a whole new concept of what actually constitutes science.   Hand in hand with the rejection of the earlier Aristotelian theories, was the repudiation of the analytical methods upon which those theories were founded.  Galileo, Newton, and others were now asking completely different questions, and arriving at significantly different answers.  One result of this new way of looking at the world was a substantially greater precision in the new explanations.  However, this precision was not the significant part of the revolution; it was the result of that revolution.  The true revolution was in the methodological procedures followed. 

   First of all, there was now little concern for classifying objects into abstract categories as a basis for building theoretical explanations.  Instead scientists were concerned with the interaction of specific variables that could be discerned to have an effect upon an event.  There was no longer a concern with the “nature” of an object, but rather with a more precise determination of the variables contributing to a situation.  An event is to be explained as the interaction of specific measurable variables (both internal and external to the object), rather than resulting from either “natural” or “unnatural” forces.  Variables rather than types formed the cornerstone of the new explanation.  This resulted from an important shift away from a concern with the abstract averages common to a group to the full consideration of a specific situation.

   Explanation assumed a whole new form.  Instead of rolling a number of balls down an inclined plane and calculating the average of the group, basing the “law of movement on an inclined plane” on that average, Galileo followed a completely different procedure.  He considered all of the variables that could influence the motion of the ball and what their effect on the ball’s motion would be.  How would the angle of the inclined plane, the shape of the ball, the surface of the plane, the medium surrounding the ball, etc. affect the movement of the ball?  By viewing each of these factors as measurable variables, Galileo could postulate how a change in one or more of these variables would affect the overall movement of the ball, making it move faster, slower --or even upwards!  The various movements were not considered natural or unnatural, but rather predictable consequences of different constellations of ball shape, plane angle, etc.  As a result, several different explanations were replaced by one single systemic explanation.  The concept of natural vs. unnatural motion became superfluous.  All motion was natural. Occam’s Razor made these two concepts irrelevant.   In addition, Galileo’s principles of motion (and the principles of those who followed him) could now be applied not only to balls rolling down an inclined plane, but any kind of motion by any kind of object.  In the place of a separate and distinct Aristotelian explanation for the movement of every single object based on the non-measurable “nature” of that object, we had a single Galilean explanation for the movement, not just of the ball but of a whole variety of objects, using a simple set of principles that predicted each behaviors as a consequence of the measurable relationship between the relevant variables.  Better and better explanations could now be proposed and tested (and accepted or rejected) by comparing how precisely they explained the motion of a variety of objects.

   This revolution in science was a fundamental consequence of by far the most unique aspect of the methodological innovations introduced by the Mechanical Revolution, i.e. the rejecting of any reference to the historically given in the formulation and determination of lawfulness.  Whether an event described by a law occurs rarely, often, or not at all is irrelevant to the validity of the law.  What becomes lawful is not the historical association between items, but rather the theoretical relationship between variables.  The historical occurrences are not lawful; instead, they are deduced by aid of the laws.  Consequently, “historical rarity is no disproof, historical regularity no proof of lawfulness” (Lewin, 1935:26). “The Newtonian achievement rendered Aristotelian patterns of explanation logically incoherent” (Wilson, 1969:298).

 How does all of this relate to explaining human behavior?  We need to ask ourelves to what extent people in their daily lives, as well as sociologists and anthropologists in their research, use concepts and explanations that are fundamentally Aristotelian?  For example, when we use the concept “mother’s instinct” to explain maternal behavior, are we attempting to explain certain behavior typically associated with women by proposing non-measurable essentialist forces considered “natural” to females?  If so, this is Aristotelian!  Similarly, if we discuss domestic violence or warfare in terms of the more aggressive “nature” of males, we are also relying on non-measurable essentialist forces attributed to the category male based on the fact that men fight wars and physical violence in our society is more typically associated with males.  Again, we are being Aristotelian. If we do adopt such approaches, then how do we explain maternal child abuse?  Do we consider these as exceptions to the rule?  If so, aren’t we saying that one form of behavior (nurturing children) is “natural” to women while the other form of behavior (child abuse) is “unnatural”?  This is Aristotelian.  If we assume that male-on-female domestic violence results because males are “naturally” more aggressive than females and that warfare is a product of male aggressiveness, then how do we explain the fact that most men do not beat their wives and children and that governments generally have to resort to the draft (which many men resist) in order to raise armies, or that lower-income males are far more likely to join the army than upper-income males?  Also, how do we explain husband abuse by wives, as well as the well-documented domestic violence that occurs in both gay and lesbian relations?  Furthermore, if these “natural” tendencies exist in all males and females, how do we explain the fact that rates of murder, domestic violence, child abuse, infanticide, etc. vary from one society to another, in the same society through time, and among different segments of a specific community?  Similarly, how do we explain the fact that the incidence of warfare varies significantly among human societies, as well as for the same society over time?  If we are going to argue that males are “naturally” more aggressive because of their higher involvement in violence, then we must also argue that African Americans are naturally more aggressive because they display a higher incidence of violent behavior as well.  Otherwise, we are being logically inconsistent.  If we are going to base the existence of natural characteristics on behavior that is typical of men and women, then we must be consistent and apply it to all behaviors that are typically associated with one sex or the other, or with different social groups.  We would need to argue, as many individuals do, that men are more naturally leaders because they fill more leadership positions, while women are naturally subservient because they have more typically been nurses rather than doctors and secretaries rather than bosses.  Similarly, we would have to claim similar natural differences between whites and blacks, as many people do.  Indeed, such Aristotelian thinking underlies all gender and ethnic stereotypes.  Such reasoning can’t, of course, explain the radical changes that have taken place in our own society in a mere 30 years.  Thus, just as the Aristotelian approach to the motion of a heavy object such as a ball cannot explain all of the variation in a ball’s motion, so also are Aristotelian concepts in the social sciences unable to explain the considerable variation that exists in human behavior.  Put very simple, a constant cannot explain a variable, be that constant the nature of a ball, maternal instinct, ethnic stereotypes, or the culture of a particular population.

    We need to abandon essentialist Aristotelian explanations in the social sciences in favor of conditional explanations that account for variation in social behavior in terms of the different material circumstances in which individuals and populations find themselves.  If, for example, we can explain differences in gender-related behavior by focusing on such quantifiable variables as the work load of men vs. women, the division of labor between men and women in the household economy, differences between men and women in control over household resources, the role of men and women in the larger political economy, the role of men vs. women in warfare, etc., without ever having to mention purported “natural” characteristics, then the principle of Occam’s Razor states that such concepts are irrelevant and should be abandoned.  They add nothing to the explanation.  They are superfluous.

    Likewise, if we can explain various social behaviors among different human societies without using the concept of culture (which is a notoriously subjective and non-measurable concept), then Occam’s Razor suggests that this concept is also superfluous and needs to be abandoned if we hope to develop systematic and testable theories of human social behavior that are able to explain and predict human behavior in any meaningful way.  When we use the concept of culture, for example, to explain the behavior of people, we are making an Aristotelian generalization?:  “Most of the people in this place behave in a certain way; it must be because they all share this non-measurable set of values and beliefs that cause them to act that way.”  But what about the people who do not behave in the way that is typical for the group?  Also, what about people who live in other societies who do the same thing?  For example, The Iroquois Indians practice both matrilocal residence and matrilineal descent?  So do the Navajo, as well as the Bemba peoples in Africa. How do we explain their behavior?  Do we attribute the Iroquois behavior to Iroquois culture, the Navajo behavior to Navajo culture and the Bemba behavior to Bemba culture, or do we see post-marital residence and kinship as part of a larger system of behavior in which different sets of ecological and economic circumstances produce different residence patterns, family structure, kinship systems, political organization, etc.?  Which gives us a better scientific understanding of the post-marital residence and kinship: having several different Aristotelian explanations based on the non-measurable cultural forces associated with each group, or a single Galilean explanation based on the relationship among a specific set of social variables from which the different forms of residence, kinship systems, political organization, etc. can be predicted?  Also, how do we explain the fact that before they acquired the horse, such Plains Indians as the Cheyenne, Sioux, Arapaho, Crow, etc. were matrilocal and matrilineal, whereas afterwards their residence and kinship became much more strongly male-oriented?  Is it necessary  --or even useful--  to talk about Cheyenne culture, Sioux culture, etc., or can we dispense with these concepts and focus instead on the impact of the horse and the gun on the role of men and women in subsistence, the ownership of property, participation in external trade relations, etc.?  Do we need a different explanation for each group, or will one systematic explanation that applies to all groups suffice?  In other words, does the concept of culture add to the explanation, or is it superfluous?

    Similarly, how useful are such essentialist Aristotelian concepts as Patriarchal and Matriarchal society?  Are these operationally definable concepts?  Don’t those who apply these concepts in their analysis of gender relations attribute specific inherent “natures” (that is, essentialist characteristics) to societies defined by these terms?  Furthermore, if so many different societies are defined as Patriarchal, then how useful is this concept?  How meaningful is it to use the same concept to describe both the U.S., (where women inherit property, participate in the labor force, comprise over 50% of students in college, medical school, law school, etc.) and the Inuit (where women do not inherit property but rather are the property of their husbands, where female infanticide is practiced, and where husbands lend their wives to other men for sexual purposes)?  How does the application of this concept help explain the much greater incidence of rape and violence against women in the U.S. than in Japan (which most individuals would describe as more patriarchal than the U.S.)?   Can we use this concept in a systematic way to explain differences in gender relations over time and in different societies, or is it, like the concept of culture, too vague and too subjective to be of any meaningful scientific value?  If it does prove unnecessary, it should, like other Aristotelian concepts, be discarded as irrelevant.

 Various patterns of social behavior occur cross-culturally with regard to ecology, kinship, gender, religion and many other aspects of human social behavior. I submit that we can better understand the cross-cultural diversity and similarity of such behavior if we approach them through the consistent application of systemic models that attempt to explain such behaviors as a consequence of underlying ecological and economic conditions, rather than through the use of vague and operationally indefinable concepts that result in logically unconnected and non-systematic explanations or in explanations that are so vague and subjective as to lack scientific credibility.



*     *     *     *     *



Of Related Interest:






Science and Anthropology






Hill's Criteria of Causation




Why Cultural Anthropology Students Should Learn

Quantitative Research Methods





A Temporary Convenience

A Critical Review of the Species Concept




*     *     *     *     *