Hi everyone! The following is my original essay on the subject stated in the topic’s title. Feel free to post any questions, comments, objections, etc. that you may have for this essay. Here is the essay itself:

All social sciences claim to be objective sciences rather than areas of philosophy because they strictly follow scientific method in their search for truth. Or at least that’s what social sciences are supposed to be about if they are to have the authority of most natural sciences. However, in reality, in the social sciences, the scientific method is hardly followed at all. For example, a hypothesis is supposed to be discarded once a single reproducible counter-fact is found. In social sciences however, a hypothesis against which there is a multitude of reproducible counter-facts is never discarded, instead, it is paraded as a sort of truth, with occasional remarks that it is not perfect because it does not work in all cases for which it was designed.

An example is in order. The theories of evolutionary psychology can often be seen in the popular press. They usually consist of various explanations of human behavior from an evolutionary standpoint. Thus, a certain conveniently chosen, human behavioral trait which is thought to be universal and hence genetic, is shown, through a few simple arguments, to increase the chances of reproduction of the individual human that exhibits such a behavior. In all this theorizing, it does not seem to matter, to our evolutionary psychologists, that cultural anthropologists have convincingly shown, time and time again, with a multitude of facts, that aside from a few simple behaviors evident at birth, such as the suckling reflex in infants, no human behavior is universal but instead can radically vary from culture to culture. Thus, aside from …, all human behavior, cultural anthropologists tell us, is learned not genetic. Therefore evolutionary theory is inapplicable for explaining any such behaviors. It is also important to note in this example that evolutionary psychologists chose only those behavioral traits which can be related to individual reproduction. All other inapplicable behaviors are disregarded even though it is usually at least implied that the same line of evolutionary thought can explain all human behaviors since it does so in non-human animals. Moreover, just because a certain behavior is universal does not mean that it is genetic. This is the case even in non-human animals. For example, female Rhesus Monkeys that grow up naturally in the wild (in a social group), readily mate with males and take good care of their offspring. However, a number of classical experiments in ethology showed that female Rhesus Monkeys raised in the lab in isolation usually refused to mate when presented with a male. Those few females that did mate and became pregnant, always either neglected or abused their infants. Thus, it is clear that even in non-human animals, all of whom (unlike humans) have many complex, definitely genetic, behaviors, a significant number of behaviors including those directly involved in reproduction, do not develop in the absence of social learning. Therefore the existence of these behaviors cannot be explained by evolutionary theory.

It is also important to note that some natural scientists sometimes fail to follow the scientific method the same way that social scientists do. For example, in ecology, there is a set of three Community Diversity Hypotheses. They are Time Hypothesis, Structural Complexity Hypothesis, and Environment/Competition Hypothesis. All three hypotheses are alternative explanations for the same phenomena – community diversity, which automatically means that if one of them is right, the other two must be wrong. However, neither one is right because all three have reproducible facts stacked against them. And yet all three are presented as being truly scientific, empirically supported hypotheses.

Misuse of statistics is also prevalent in both natural and social sciences. Statistical data can only show correlation between two or more variables; it cannot show any empirical connection between these variables, yet alone what effect these variables have on each other.

An example is in order. Perhaps the most amusing, invented example from medical science is that of Japanese men and smoking. A truly random sample of Japanese men will contain more smokers than an equally random sample of the same size, of men from any other country. However the men from Japanese sample will, on average, be older when they die than the men from the sample from a different country. Conclusion: smoking prolongs life probably by strengthening the health of the smoker. Another amusing, invented example comes from criminology. If one compares the total number of cops and the total number of homicides in settlements of different sizes, one will see that as the total number of cops in the settlement increases so does the total number of homicides. The same pattern can generally be seen for all other types of crime. Conclusion: cops cause crime, without them our towns and cities would be crime free. Thus, the use of statistics as described above, is scientifically useless and is nothing more than a skillful creation of false facts.

Even when statistical data presents only a single variable, it is seldom reliable enough to support a hypothesis. A good example comes from the article “The new face of AIDS” from The Economist magazine. The article can be found at http://www.neww.org/en/news/print/1,534,1.html. In that article there is an attempt to show that a high prevalence of rape greatly contributes to the spread of HIV. To prove such an assertion the following statistical data from developing countries, where the rate of HIV infection is many times higher than in developed countries, is given: “In rural Peru, for example, 24% of young women say they lost their virginity to a rapist. In a recent national survey in South Africa, 10% of sexually experienced young women said they had been raped.” Such data however cannot possibly be deemed to be supportive of the hypothesis. First, it is well known that only those who have very strong feelings about the matter of the survey will respond to the survey. All others, who often outnumber the actual respondents, do not respond at all because they have little interest in the matter. Second, in cases such as this one, the matter is too sensitive for many people, so they do not respond or respond incorrectly. Third, those who do respond, can respond incorrectly to give themselves an ego boost (e.g. In a survey on exercise frequency, those people who do not exercise at all, might claim that they exercise “at least 3 times a week” in order to avoid appearing “losers”, even though the survey is anonymous.) Consequently, those young Peruvian women that responded to the survey were most likely a minority of young rural Peruvian women, and some of their responses were most likely incorrect. The same goes for South African respondents. Fourth, Peru contains a very large population of young urban women who were not included in this survey, even though the urban crime rate is always higher than the rural one. Hence, urban women have a higher opportunity of being raped. However, due to their more liberal morals, young urban women are more likely to lose their virginity to a boyfriend or an intimate partner rather than a rapist. Fifth, it has been claimed that in North America one in every 4 female college students gets raped at least once during her time in college. However the rate of HIV infection among North American women is tens of times smaller than that in South Africa and many other African countries. Thus, if a true social science is to be built, the use of single variable statistical data, for supporting any hypotheses, must be avoided.

Yet another problem prevalent in social sciences has to do with the scientific concept of the definition. Every term must be rigorously defined in order to make sure that everyone understands what the term in question stands for, what the sentences, which use it, mean, and, most importantly, to make the hypotheses, which involve the subject matter which it defines, testable. However, in social sciences these requirements are often not met. For example in cultural anthropology and sociology there is a theoretical orientation known as structural functionalism. The basic premise of this theoretical orientation is that “Culture exists for the benefit of society.”

Even though every introductory textbook in cultural anthropology attempts to give a good definition of the term culture (more advanced textbooks don’t bother defining culture at all), it only succeeds in giving a general idea about the subject matter of the discipline while completely failing to give a rigorous, scientifically useful definition, which would make every hypothesis about culture, testable, provided all other terms within the given hypothesis are also rigorously defined. Thus, a popular definition of culture is “everything people have, think, and do.” As great as this definition may seem, it says nothing about what culture really is, it only defines its boundaries. A rigorous, scientifically useful, definition of culture would have to give a precise answer to questions like the following: Is culture an independent entity which exists on its own and determines human behavior? And if so, is it modified by people? Or if not, is it created by people? Moreover in order to give a rigorous answer to these two questions, it is first necessary to answer two other questions about the nature of culture: Is culture a single entity all parts of which work together towards a common goal? Or is it a mixture of competing ideas, deeds, and man made objects? All these concepts are probably illustrated best, by looking at the analogy found in ecology. The physical environment definitely exits on its own, independent of people, and it definitely determines many human behaviors. However, it is often modified by people. Moreover, the physical environment is a mixture of competing organisms, natural forces, and inert matter. However, it is conceivable that a philosopher may see it as a single entity all parts of which work together towards a common goal, such as the creation of superior organisms through natural selection.

The term benefit is also important to define because it is very subjective. What is beneficial to society in the eyes of one individual is harmful in the eyes of the other. Thus, the hypothesis: “Culture exists for the benefit of society.” is neither clear nor testable, even though it is often presented as being an exact opposite and empirical evidence is given to support it.

Another, wider known, example involves the definition/clarification of the term/concept gender identity. The problem stems from the fact that the term is never defined even though it is freely used in at least two hypotheses which happen to contradict each other. They are “Gender identity is learned.” and “Gender identity is genetic.” Proponents of either hypothesis seem to always assume that their hypothesis is clearly understood. So instead of defining anything in it, they move on to overwhelming the reader with empirical evidence. However, it is never clear how the empirical evidence, which is being presented, is supposed to support the hypothesis. This should not come as a surprise because the key term in the hypothesis, gender identity, was never defined.

A Case Study of a Failed Attempt to Reach Scientific Conclusions:

Scholars of sexual selection theory often claim that the extent to which secondary sexual characteristics of a particular male are developed is directly proportional to the average level of testosterone in his blood. In other words, a peacock with a larger, brighter tail must have a higher average level of testosterone in his blood than a peacock with a smaller, darker tail. Consequently, one of the answers to the question why a peahen always finds the peacock with the largest, brightest tail most attractive, is that testosterone weakens the immune system; hence a peacock with the largest, brightest tail must be genetically very healthy, and hence the best mate, since he obviously remains healthy in spite of his immune system being under a constant attack by testosterone. This hypothesis as a whole, however, is incorrect for at least 3 reasons.

First, the extent to which secondary sexual characteristics of a particular male are developed is not directly proportional to the average level of testosterone in his blood. In fact, the extent to which any particular secondary sexual characteristic is developed is more a matter of the extent to which the relevant tissues respond to testosterone, rather than being a matter of the absolute level of testosterone in the blood. Consequently, if we take two males with identical levels of blood testosterone, one of them can easily supersede the other in the development of any one secondary sexual characteristic. Take muscular development among humans. An average guy who goes regularly to the gym, is a weakling compared to Arnold Schwarzenegger, who stated that he did not use steroids to build his muscles (ref.: http://web.archive.org/web/20031008172601/http:/hjem.get2net.dk/JamesBond/www/artikler/steroidemisbrug/arnoldandsteroids.htm). Thus, if we assume that Arnold Schwarzenegger had an average level of testosterone, then it is clear that the effects of testosterone on his muscle tissue, namely increased protein synthesis and reduced recovery time, were many times greater than that for an average guy. Of course it is possible that Arnold’s natural level of testosterone was several times that of average. In that case it is at least clear that his high testosterone levels, while causing great muscular development, failed to cause a significant growth of bodily hair, such as chest hair (another secondary sexual characteristic). Photos of him in the gym with fully grown (after a not so recent contest) armpit hair show no evidence of chest hair. And of course as the above given reference on Arnold’s steroid use suggests, Arnold could have lied about not using steroids to build muscle mass. This however would not mean that testosterone levels within the natural range cannot cause massive muscle growth. In the words of Lee Labrada “Steve Reeves, probably the best-known champion up to 1965, had 19.75” calves, before steroids had even arrived on the scene” (ref.: http://labrada.com/article_detail.php?acode=123). Arnold’s calves were 20”. In all other measurements Steve Reeves was also close to Arnold. Let’s take another extreme and look at the author of the present article. He is in his mid-twenties, and compared to Arnold, he is remarkably hairy. Hence, it is clear that the growth-stimulating effect of testosterone on his bodily hair is many times greater than that for Arnold. However, testosterone does not seem to have any post workout anabolic effect on the muscle tissue of the present author. In fact, performing a regular workout of an average guy that goes regularly to the gym, leaves him with severe muscle cramps that take a week to disappear. Moreover, in spite of regular workouts (suited to his abilities) the rate of growth of his muscle tissue is next to zero. Finally, African-American men are known to have testosterone levels nearly double those of European-American men, however, compared to European-Americans, they have hardly any bodily hair. They even lack hair on their forearms.

Second, even artificially induced, very high levels of testosterone, unlike high levels of stress hormone cortisol, do not weaken the immune system. In fact testosterone, at the very least, prevents cortisol from also attacking muscle tissue.

Third, even if testosterone was harmful to the immune system, a male with high testosterone levels would not make a good mate even if he appears to be healthy in spite of his condition, because all conditions, such as high testosterone levels, are potentially inheritable. Hence mating with such a male would very likely result in offspring with heightened testosterone levels and hence potentially weaker immune systems.

The above case study shows that scientists sometimes either make conclusions without sufficient supporting evidence or simply fail to consider known counter-evidence (first problem of the analyzed hypothesis). Also, scientists sometimes either invent supporting evidence for their hypotheses, or disregard known counter-evidence (second problem of the analyzed hypothesis). Finally, scientists sometimes either ignore some of the evidence pertinent to their arguments or simply fail to consider it (third problem of the analyzed hypothesis).