In this critical analysis of testing in the sciences, we will examine the various issues for test in the human and natural sciences. We will examine the theories of Searle, Taylor and Popper. In this analysis, we will strive to uncover which of the theories, which they represent; best explains the Philosophy of Social Science. The claims made by Searle and Taylor for studying natural and social science, will be compared with Popper’s Account for Testing in the Social Sciences, in order to examine the various methods currently being employed in studying the Philosophy of Social Science.
Natural Science is divided into two distinct categories: Physical Science and Life Science. Physical Science includes the study of physics, as well as Atomic Science. Life Science is the study of various species and their roles in the economy of nature. Social Science involves utilizing a method of study to examine factors of human life; including social, economic and human development. Natural Science can be studied according to the scientific method and scientists would develop a hypothesis and then test, using various apparatus and experimental strategies. Social Scientists might limit their study to examining the social interaction of human beings. Therefore, Social Science cannot be studied using the same methods as Natural Science. Social Science examines human beings and social life, which is a domain that contains far too many variables to be calculable. Natural Science can be studied and experimental results can be empirically tested and retested, successfully. The study and testing of experimental results is possible because Natural Science examines substance(s), which scientists proclaim to be calculable.
We will begin our discussion by analyzing the claims made by Searle for studying Social Science. Searle presents four points about human science:
Searle’s first point is “the that explanations of human behavior are casual…but that such explanations whether of individual behavior or collective behavior, make references to a special kind of causation – they require a certain form of mental, or as I like to call it, “intentional” causation.
(Searle, 1991: 35)
We might perceive Searle’s first point to be that not only is human behavior causal, but so are the explanations that we might ascribe to human behavior. This claim might require the additional support of the supposition of collective intentionality, which Searle discusses in his book, On the Construction of Social Reality. Searle also appears to claim that such explanations of individual or collective behavior make reference to a special kind of causation. Before we examine the remainder of Searle’s first point, we must first examine this aspect of his claim very closely. If there is a special kind of causation to which explanations of individual and collective behavior make reference, then there might be more than one type of causation. Hopefully, Searle will explain the various types of causation that exist. Proceeding with Searle’s first point, it appears that he claims that the special causation, to which explanations of individual and collective behavior make reference, requires a certain form of mental state, which he refers to as “intentional” causation.
This might suggest that the mental – that which involves the mind – produces “intentional” causation. It also forces us to wonder if there are other kinds of causation (which are not mental and which are not intentional), yet are still forms of causation nonetheless. A possibility exists that perhaps these other kinds of causation are natural processes. Nevertheless, “intentional” causation which occurs as a result of the mental (or in the mind) would be the product of human will to cause something to exist or have an existence, even if only occurring in one individual’s mind, or possibly in the mind of the collective.
Searle’s point might be that the explanations of human behavior occur in the mind of the social scientist(s) who study the behavior. This varies from natural scientific study, in which the explanation might only be perceived in the mind of the scientist, while evidence is present in or on the apparatus. Searle might be suggesting that Social Science is not testable in the same way as Natural Science, because social science experiments are conducting by observing and examine the behavior of test subjects. After the observations are made, the data is then analyzed and conclusions for the experiment are produced according to the methods implemented the social scientist(s). The difference between this kind of experiment, which is done by the social scientist and one performed by a natural scientist, is that a scientist of natural phenomena tests and retests his experimental data; by using various apparatus and his conclusion is based on physically evident results. If conducted according to the scientific method of conducting an experiment, likely, other scientists will be able to duplicate the experiment and observe similar results.
For the social scientist, the results of his experiment are merely the product of his mind, while examining the data. It is unlikely that other social scientists will find similar conclusions if they duplicate the same experiment. Also, while test subjects, in social science, have a physical existence and are comprised of substance) according to natural science), their behavior is not tested using a physical apparatus in an absolutely controlled environment) as in the natural sciences.
The second point I want to make is that there is a class of social facts having certain logical features that make them quite unlike the phenomena of physics and chemistry. In the case of phenomena such as marriage, money, divorce, elections, buying and selling, hiring, firing, wars and revolutions, the phenomena are – to speak vaguely at this stage – permeated with mental components; furthermore, the facts in question are self-referential in an odd way because they can only be the facts they are if the people involved think that they are those facts.
Searle’s second point is extremely diverse, therefore, we will examine it in three parts: Firstly, he claims that there is a class of social facts (but not necessarily all social facts), which by having certain logical features, make them quite unlike the phenomena of physics and chemistry. There is a class of social facts that are applicable, yet there is also a class of social facts to which Searle’s second point would not apply. The remaining claims (referring back to his second point) suggest that some social facts that humans have labeled are tangible and tradable, while other so-called social facts are merely institutional facts, such as marriage and divorce. The social fact of money (which is tangible) is a social fact that is tangible as currency, yet money is also an institutional fact, in the case of credit and debt economics.
Most importantly, in Searle’s claim regarding the phenomena, is that they are permeated with mental components. This suggests that these phenomena, while they are observable in the physical world, are items that have been labeled and defined as institutional facts and social facts. Towards the end of his claim, he explains how these social and institutional facts are maintained. Searle clarifies that the facts in question are self-referential in an odd way, because they can only be the facts that they are, only of the people involved think that they are those facts. This claim seems to be very persuasive.
If the facts that Searle presents us with in his second point are self-referential, then it would lead us to believe that if we did not refer to these facts at all, then they would not exist. It would, however, not be sufficient for us to deny, ignore or fail to acknowledge the existence of these facts. Collectively, we would need to abstain from using these facts, and only then, they might cease to be recognized as facts. But Searle does not explicitly discuss this possibility. He does divulge though, that these ‘facts’ can only be the facts that seem to be, if the people involved think that they are those facts. Searle might be demonstrating that the existence of these facts does not necessarily depend on everyone accepting these facts, or affirming their purpose or existence. Instead, he illustrates these facts, to which he is referring, can be seen as ‘facts’ only by the people involved within the institutional structure where these facts are maintained. Whoever is not involved within the institutional structure and does not make use of these social facts, would not affect these ‘facts’ whatsoever.
Most compelling, of Searle’s arguments, is the differentiation between social and institutional facts and natural scientific facts (such as physics and chemistry). Natural scientific facts do not need to be permeated with mental components. They exist and operate independently, within the economy of nature, regardless of human acknowledgement or definition of them. This persuades us to believe that natural facts have an existence independent from social reality. Perhaps this notion will help us assess the testability of natural and social facts. Social facts are easier to test, simply because they exist within institutional structures that have been defined by human beings. Therefore, the social and institutional facts that exist can only be understood by people living within that particular institutional structure (who have previously defined these facts as ‘facts’). They are then referred to as either social facts or institutional facts. Natural phenomena are more difficult to test than social phenomena, because testing it involves the introduction of an apparatus and other items that might affect the phenomena being tested. The introduction of other an apparatus and other items might affect the phenomena to such a degree that the test would be invalid according to the scientific uncertainty principle as proclaimed by Heisenberg.
A third distinction between the social and natural sciences is a direct consequence of the nature of intentional causation. The prepositional content given by the theorists in the explanation of the behavior must be identical with the prepositional content in the actual mind of the agent or agents whose behavior is being explained; otherwise, the behavior is not properly explained.
(Searle, 1991: 337)
It now appears that the problem of testability in the natural and human sciences is more complicated than we had originally believed. Rather than the testability of social science being more reliable than natural science, it might in fact be more unreliable and more difficult to understand. This problem is emphasized by Searle’s claim regarding the nature of intentional causation. If “the prepositional content given by the theorists in the explanation of the behavior needs to be identical with the preposition content in the actual mind of the agent for the behavior to be properly explained, then where this were not the case, the behavior would not be accurately perceived.” (Searle, 1991: 337).
A fourth feature of social phenomena is also a consequence of its intentionality. It is this: For this reason, then name of the phenomenon is often partly constitutive of the phenomenon so name. The phenomena are completely unlike, for example, such physical phenomena as gravity or kinetic energy and such biological phenomena as diseases or hereditary traits. Whether or not something is a certain disease or whether or not certain relations of gravitational attraction exist between two entities is a fact that is completely independent of how it is represented. They exist independently of what anybody thinks about them. But in the case of social facts, the beliefs and the terms that people use are partly constitutive of the facts.
(Searle, 1991: 339)
In Searle’s fourth elaboration on social phenomena, he attests to what might be the greatest problem for the testability of social science. He then further demonstrates how and why natural science and the phenomena that it studies are testable in a fashion that social facts are not. Whether or not a fact in natural science is represented appropriately, regardless, the conception of a fact does not negate its testable qualities. Even if a natural fact is misinterpreted in explanation, the fact can still be studied and tested by another scientist, thereby achieving similar results. On the contrary, in social science, where a social fact is misinterpreted as a different social fact, the actual portrayed fact will be the one that will be tested and perceived.
Just as technical jargon (usually Latin) is employed in the natural sciences, technically specific language is the main tool used for studying social science. If the terminology is flawed and categorical imperatives are not clearly established in the social sciences, then the experiment will be faulty. In contrast, even if the terminology in natural science is inaccurate, the experiment would still be useful, as long as the phenomena being studied is correctly portrayed through visual means (or can be derived by scientific deduction).
Let us now examine the claims made by Taylor (as explained by Staunch) and determine whether or not they are in agreement with Searle’s viewpoints.
Where the natural sciences were seen to be using explanation of phenomena based on theoretical laws formulated as a result of empirical observation of behavioral regularities, the human sciences were held to be involved in the attempt to understand the meaning of the behavior that composes the phenomena of its domain.
(Searle, 1992: 338)
According to Staunch, the natural sciences employs explanations of phenomena based on theoretical laws derived from empirical observation of behavioral regularities, whereas, human sciences involved in understanding the meaning of the behavior that composes the phenomena of human beings. Human science is directed at applying theoretical laws to observations of the behavior of test subjects. The difference between the two approaches taken by these fields of study is very obvious. Natural scientists examine data gained from conducting experiments and derive laws from repeatedly testing the data. Social scientists, on the other hand, continue to examine human interaction and human institutions, in an effort to understand the relationship of human to their environment and the reasons for human reaction to socially constructed phenomena.
“Thus the distinction that Taylor draws between the natural and social sciences is based on the ontological difference, and this makes Taylor’s position quite close to Dilthey’s who also sees the difference between the human and natural sciences as difference of ontology” (Staunch, 1992: 342). If the difference between these two science in ontological, then we might reasonably question what is meant by the term ‘ontology’. If ontology refers to the origination and the source of these sciences (which are different for the social and natural sciences), then the ways that these sciences evolved are not necessarily relevant. Instead, the ontology of Social Science (according to Searle), is derived from collective intentionality and intentional causation.
Where the differences between these sciences are evident, these differences highlight the Social Sciences as a being easier to tracer to its roots. On the other hand, the phenomena in natural science can be tested and these tests might provide result that natural scientists could readily use to better understand the natural world. Natural scientists can also derive theoretical laws for the world by testing various phenomena. The study of phenomena in the Social Sciences does not result in the derivation of theoretical laws, because the phenomena observed are the actual objects of study. Humans, as well, as their mental states and social creations, are the phenomena of social science. The phenomena can be ontologically traced back to the collective intentional imposition of participants, belonging to the society where these social facts exist (Searle, 1995).
“Rorty, who has subsequently argued that the only distinction worth making between thee sciences is a moral one that does not depend on either a methodological or an ontological basis” (Staunch, 1992: 340). If Rorty’s reasoning is correct, the it would not be worth distinguishing between natural science and social science (the only distinction worth making would be a moral one). Rorty emphasizes that methodology and ontology are not factors when establishing the moral differences between these sciences. His assumptions might be correct, but his argument lacks evidence. Despite the fact that there appears to be no methodological differences between these sciences, they should still be studied using the same methods, making use of the relevant apparatus. Yet, bear in mind that the apparatus in social science is language and social constructs. The methodology, therefore, would not be exactly the same for both sciences. The ontology of these sciences, however, might be very similar. Social Science and social facts could have developed from natural processes that have been developed in human beings; thereby influencing various characteristics within them. Natural science might be the ontological source of social science; whereas Social scientists are more concerned with studying the behavior of human beings in the social world. This is far more significant than analyzing how the social world might have been biologically derived from natural phenomena.
“Thus social scientists carry an additional, or at least a different, responsibility relative to natural scientists. For social theory plays a creative role in the social order and continuation of human practices and institutions in a way that natural science doe not” (Staunch, 1992: 342). Social scientists have an additional responsibility relative to natural scientists. This responsibility is twofold; as well as studying human being as products of natural phenomena within their natural and social worlds, social scientists must examine the socially constructed world and human relationships within it towards and it. Social scientists’ carry a burden with them that is unique, yet respectably equal to the pursuit of natural scientists’, in discovering the processes of development in the world.
“Consequently, the social sciences are involved in a double hermeneutical practice because they must interpret what is already an interpretation” (Staunch, 1992: 347).
Let us refocus on the claims made by Searle (specifically his second point), and determine how Popper’s claims either agrees with or refutes Searle’s claims.
Popper’s theory of explanation in the social sciences is intended to avoid the errors of those social theorists who have either ascribed too much reality to social groups (social classes, the nation-state) or have, contrarily attempted to reduce all phenomena (law, money) to the psychology of individuals. On the one hand Popper has asserted that it is wrong – both factually and morally – to ascribe to social whole a will of their own. And it is a mistake to regard it as the taks of social sciences to prophesy the life-cycle of these super-individuals…Popper asserts because our social action can have effects quite the contrary of our intentions.
Searle’s second point exemplifies ideas that are exactly the opposite of what Popper claims should not be done. That is ascribing too much reality to social groups and reducing all phenomena (i.e. law, money) to the psychology of individuals. But Popper is not only declaring that this approach is loaded with errors. He also claims that it is wrong to ascribe a will to their own social wholes. “This view must be mistaken, Popper asserts, because our social actions can have effects quite the contrary of our intentions” (Berkson, 1989: 61). Searle argues the contrary viewpoint to Poppper. Searle argues the contrary viewpoint to Popper. Searle claims that facts, whether social or institutional, are self-referential and they can be only the facts that they are if the people involved think that they are those facts (Searle, 1991:335-336). We cannot be certain which viewpoint is actually correct, when we contrast and compare Searle to Popper. Yet, it is evident that there is contrasting evidence in the social sciences that is persuasive and compelling, which would allows us to accept two claims as competing theories (neither or which being more persuasive than the other). This is similar to the theories in natural science. Theories simply remain theories in natural science and do not become scientific laws. This mainly occurs for two reason: Either empirical studies cannot be replicated to achieve the identical results of the original study or because the contemporary theory conflicts with an already established (or equally compelling) theory.
In Popper’s closing remarks, it is explicitly demonstrated that he drastically refutes the position taken by Searle. “Popper asserts because our social action can have effects quite the contrary of our intentions” (Berkson, 1989: 160). Searle agues the opposite side of this claim:
Social facts differ from natural facts in that they contain mental representations. But they differ from other facts in that the mental representation have the element of self referentiality that I was just attempting to adumbrate. The thin is only what it is if people think that it is what it is.
(Searle, 1991: 341)
Searle also claims that, “in the case of social facts, the beliefs and the terms that people use are partly constitutive of the facts” (Searle, 1991: 339). He then proceeded to argue the following:
Intentional causation differs in an important respect from the sorts of causal phenomena that we are familiar with when we discuss things such as gravitation or nuclear forces, for intentional causation is that form of causation involving mental state in virtue o actual content of the mental states.
(Searle, 1991: 337)
Popper might be correct in his assertion that ‘our social actions can have effects quite the contrary of our intentions’. If Popper’s assertion is correct, then it is difficult to determine what we can make of Searle’s claims regarding intentional causation or collective intentionality. Searle’s claim represents the norm social actions, and in contrast, Popper’s position asserts that there are exceptions to the rules of social action. Social actions, like much of social science, might therefore be theoretical and not empirical. This might be the determining factor that distinguishes it from natural science.
In the physical sciences, theories are routinely testable, and the value of testability is unquestioned. In the social sciences, however, testability is not the general norm, and in fact the social scientists who want to produce testable theories of society faces a dilemma.
(Berkson, 1989, 157)
In summary, Searle proclaims that, “indeed there is a sense in which, for the most part the social sciences have not advanced theoretically beyond a kind of systematized common sense” (Searle, 1991: 336).
Berkson, W. (1989). Testability in the Social Sciences. Philosophy of the Social Sciences, 19, 157-171.
Searle, J.R. (1991). Intentionalistic Explanation in the Social Sciences. Philosophy of the Social Sciences, 21, 332, 344.
Searle, J.R. (1995), The Construction of Social Reality. New York: The Free Press.
Staunch, M. (1992). Natural Science, Social Science, and Democratic Practice: Some Political Implication of the Distinction Between the Natural and Human Sciences. Philosophy of the Social Sciences, 22, 337-356.