When psychology instructors tell students that they have raised “an interesting empirical question,” these instructors do not typically mean that students have engaged in an empirical ideology. Indeed, it is likely these instructors mean the opposite of engaging in an ideology because they consider empiricism a kind of scientific method for mapping “objective” reality and avoiding ideologies altogether. Often, in fact, the term “empirical” is used as a synonym in psychology for unbiased or scientific.
We describe this common student/instructor exchange in psychology because it exemplifies a prominent misconception in the discipline – that empiricism is a kind of transparent window that reveals the objective truth of the world. As we will explain, however, the philosophy or epistemology of empiricism is anything but transparent because it has its own values and assumptions. In fact, these values and assumptions could be viewed as a kind of “disguised ideology.” According to Richard Bernsten (1976), a disguised ideology occurs when “value biases have been confused with factual descriptions in explanatory social science” (p. 104). As we will see, empiricism has clear “value biases” that provide a privileging of certain aspects of our experience over others, yet these values are often presented to students as the “facts” or “logic” of science.
From this perspective, an awareness of this ideology is vital to the province of a critical psychologist because empiricism is both a “dominant account of psychology” and used in the “service of power” (Parker, 1999, p. 11). Few would question its dominance, as virtually all the prominent research methods texts evidence (e.g., Dyer, 2006; Mitchell & Jolley, 2007; Schweigert, 2006; cf. Slife, Reber, & Faulconer, 2012). But citing its power is provocative because methods are rarely viewed as having political or economic implications. Still, one only has to consider the central role of empiricist values and assumptions in evidence-based practices to realize the economic power of these therapeutic practices in insurance reimbursement. Clearly, the power implications of this disguised ideology could be mightily important to critical psychology specifically and the social sciences more generally.
Empiricism is the philosophy or epistemology that our knowing and learning is primarily derived from our experience. However, this essay examines how empiricism has historically varied in what is viewed as valid forms of experience, from narrower forms that embrace only observable experience to broader forms that allow more experiences than merely visual experiences, such as emotional and spiritual experiences. Interestingly, only the most narrow version of empiricism is typically described in psychology’s research method texts (e.g., Dyer, 2006; Mitchell & Jolley, 2007; Schweigert, 2006). As we will also see, this narrowed version continues even when operational definitions are taken into account.
Empiricism; research; science; disguised ideology; operationalism; prejudice
The philosophy of empiricism has a rather long history, at least as far back as the ancient philosophies of Democritus and Aristotle. Even the Middle Ages had strong champions of empiricism. William of Ockham, for instance, favored sensory contact with objects of experience, what he called intuitive cognition, in the fourteenth century. However, the most influential proponents of empiricism for psychologists were the seventeenth and eighteenth century British philosophers John Locke and David Hume who built entire philosophies of mind around the notion that knowledge is derived from sensory experiences (Leahey, 2004; Rychlak, 1981; Slife & Williams, 1995).
This Lockean/Ockhamian tradition is the narrower brand of empiricism, exemplified most prominently in many of psychology’s quantitative methods. Sometimes known as naïve empiricism (e.g., Strong, 1991), its doctrine that only the observable should count as knowledge is currently so taken for granted in psychology that many students find it odd to consider it a philosophy; it is for them the way science is conducted. As Mitchell and Jolley (2007) stated in their research method text: “To avoid being swept away by either unfounded speculations or biased perceptions, scientists tie their beliefs to concrete, observable, physical evidence that both independent observers and skeptics can double-check” (p. 4).
Challenges to this narrow or “naïve” empiricism originated from several historical quarters. Perhaps the two most significant were arguments that this tradition of empiricism discounted or omitted important interpretive and social elements. The eighteenth century philosopher Immanuel Kant (1783/1996) is probably most noted for asserting the interpretive element of experience while Karl Marx is distinguished for contending the importance of social factors in experience. As Kant put it, the British empiricist Hume “interrupted my dogmatic slumber, and gave my investigations in the field of speculative philosophy quite a new direction” (Kant, p. 7). With Kant’s more rationalist leanings, his philosophical investigations moved in a direction that challenged the notion that knowledge comes directly into our minds from sensory experience. He eventually contended, instead, that our minds are naturally prepared to organize and give meaning to experience. In other words, Kant held that at least some knowledge results from mental activity that is logically prior to experience. From this perspective, the knower makes sense of the world through organization and selective attention. The world does not make sense of itself.
The philosophy of (Marx & Engels, 1848/2002), on the other hand, could be viewed as indirectly challenging the narrow brand of empiricism for its relative neglect of other factors, including social and ethical factors (Taylor, 1966). First, from Marx’s perspective, giving so much credence to the merely observable can lead to the underestimation of how much this observable depends on the culture or society. In a modern investigation of child abuse, for instance, conventional empiricists can too easily overlook how much their society shapes their understandings of what counts for child abuse. As Gergen (2009) has demonstrated, these understandings can differ not only across cultures but also across time within the same culture.
Second, Marx rejects the traditional empirical distinction between fact and value. Many conventional empiricists, for example, tend to view their data as representing the objective and value-free reality of the world – the so-called facts (Slife & Williams, 1995). Yet, this view discounts the importance of a number of value-laden investigator decisions, including the value of the topic investigated, the value of the particular method design selected, and the value of the particular interpretation made of the data. As Slife and Williams describe, the data themselves, past or present, dictate none of these decisions. These decisions are the value-laden conditions of the data, not their result.
Challenges such as those of Kant and Marx have led to a broader notion of experience and thus empiricism in many qualitative approaches to research. The philosophy of one of psychology’s “parents,” William James, perhaps best typifies this broader tradition of empiricism. This tradition embraces more than merely sensory experiences. It assumes that humans can have knowledge of their thoughts, feelings, meanings, and even spiritual experiences. James’s (1902/1982) classic book, The Varieties of Religious Experience, is an example of a book that attempts to further knowledge about spiritual experiences in this more inclusive sense of empiricism. Although experiences of our feelings, meanings, and thoughts do not fall on our retinas, they are nevertheless central to a broader understanding of experience and we depend on our knowledge of them everyday (Slife & Melling, 2009). James’s radical empiricism (1912/1996) is “radical” in this sense because of this experiential inclusivity.
Many phenomenologists and hermeneuticists are interested in this broader understanding of empiricism, especially if empirical experience includes meanings (Packer, 2011). The reading of books is an example of how meanings do not fall on our retinas. Although the printed words on a page are clearly observable, and thus fall on our retinas, the relation among these printed words, which is required to understand the meaning of the story, is not strictly observable. In fact, the relations among almost any items or things, including interpersonal relations, are not strictly observable (Slife & Wiggins, 2009). They are meanings that are experienced in the broad Jamesian sense of empiricism but they are not publicly observable in the narrower sense advocated in most psychological methods texts (e.g., Dyer, 2006; Mitchell & Jolley, 2007; Schweigert, 2006). Yet, people seem able to comprehend and compare understandings of these meanings as forms of knowledge (Slife & Melling, 2012).
The realization that many important psychological phenomena are not strictly observable in the narrow sense – including not only emotions, spiritual experiences, relationships, and meanings, as we have just described but also attitudes, memories, and motivations – led historically to important empirical method developments perhaps most notably that of operationalization. Indeed, most psychological texts on research methods consider operationalization a required step in formulating studies in psychology, especially when the topic under investigation is not itself publicly observable (e.g., Dyer, 2006; Mitchell & Jolley, 2007; Schweigert, 2006). Historians of psychology, such as Viney and King (2003), have credited the physicist Percy Bridgman with “set[ting] forth the principles of operationalism” (p. 302) in his classic book (1927) The Logic of Modern Physics. However, Bridgman was also one of the first to debate operationalization’s usefulness to psychology (Holton, 2005; Walter, 1990). We examine aspects of this debate after first describing operationism’s connection to modern empiricism.
Operationalism’s intimate relationship with the narrower sense of empiricism is probably best understood through a simple example. Although the authors of this essay can claim to love their partners, this love, whether an emotion or a relationship, is not strictly observable (see explanation above). This situation leads quantitative researchers who are interested in studying love to “operationalize” love in terms of observable behaviors. In an important sense, these researchers are attempting to translate the unobservable into the observable so that the topic can be investigated with the narrow version of empiricism. Typically, these researchers assume that the operationalization is a manifestation of the unobservable topic under consideration. With the example of love, this translation might mean considering love to manifest hugs and/or kisses. In others words, if the present authors truly love their partners, hugs and kisses should be manifested accordingly.
As logical as this method practice may seem, critics have noted several problems that directly involve the narrowed meaning of empiricism (cf. Chang, 2009; Leahey, 2001; Slife, Wiggins, & Graham, 2005). First, hugs and kisses are not necessarily connected to love. Hugs and kisses can occur without love, and love can occur without hugs and kisses. In this sense, knowledge of hugs and kisses, which could itself be valuable, should not be considered knowledge of love. Operationalizations, for this reason, are not necessarily identical with and may not be related at all to the construct or topic being operationalized, though this problem is rarely discussed in psychological research that uses operationalizations. Even biological operationalizations, such as fMRI scans of human brains, are not identical to the human brains they attempt to measure (Bub, 2000; Fenton, Meynell, & Baylis, 2009; Tovino, 2007). Like all operationalizations, these “scans” selectively attend to or emphasize some parts of the topic under investigation and ignore or deemphasize others.
Critics of the method practice of operationalization have also pointed to a second problem: operationalization prevents us from knowing, at least in the narrow empirical sense, the relation between the unobservable topic of interest, such as love, and the observable operationalization, such as hugs and kisses (cf. Slife et al., 2005). The relation between the two, the betweenness of the observable and unobservable, is not itself observable. In other words, we cannot empirically check the validity of operationalizations, such as how closely they represent or manifest the topic under investigation, because this relation is not itself knowable, at least from the narrow and conventional psychological meaning of empiricism.
Even a “convergence” of multiple operationalizations (Grace, 2001), where one unobservable topic of interest (e.g., love) is converged upon by several operationalizations at once (e.g., hugs, kisses, smiles), would not necessarily overcome these problems. As discussed, the operationalized relationship of one unobservable topic of interest to its operationalization is not itself empirically knowable (in the narrow sense). Consequently, adding other nonempirically knowable relationships in a convergence of operationalizations could compound rather than resolve the problems. The dominance of the narrow meaning of empiricism, and thus the prominence of operationalization as a method practice, means that psychology could be filled with studies of operationalizations that have no necessary or knowable connections to the topics of original interest (cf., Slife & Melling, 2012).
Perhaps more important, from the perspective of a critical psychologist, is the possibility that a narrow or “naïve” version of empiricism has become the international standard for investigation in psychology with all the ideological prejudices that such a philosophy implies. As Gadamer has noted (e.g., 1993), all ideologies, including all the variations on empiricism, have implicit prejudices, i.e., ways in which the ideologies reveal and conceal certain aspects of the experienced world. We mentioned at the outset the unfortunate myth in some parts of psychology that empiricism does not involve values and biases and thus prejudices. Indeed, many empiricists would claim to strive to eliminate all biases, values, and prejudices. They would claim to discover the objective world by clearing away, as much as possible, the subjectivity (and thus prejudices) of the researchers through the scientific method (e.g., control groups, experimental manipulation).
However, Gadamer (1993) and other critics of this claim view it as another manifestation of implicit prejudice, what he calls the “prejudice against prejudice” (p. 273). This is the prejudice that biases are bad, itself a type of value, i.e., the value of wanting to be value-free. These critics note that any epistemology or philosophy that guides knowledge advancement, such as empiricism, must guide that advancement by being “biased” in some sense about what matters (e.g., observables) and does not matter in science. Empirical researchers, however, rarely admit these prejudices explicitly nor are the prejudices always consciously held (Slife & Williams, 1995). Rather, many researchers are taught these prejudices in their methods training often without the recognition that they are values or biases.
This training means, for example, that when Western psychologists teach “the scientific method” to their Eastern psychological colleagues, they are simultaneously teaching an often unidentified set of philosophical assumptions, many of them singularly Western prejudices. The empirical epistemology, especially when it is viewed as synonymous with the scientific method, becomes a kind of Trojan horse or “disguised ideology” of Western imperialism. This situation was brought home to the senior author of this article when he collaborated with Chinese psychologists who wanted to develop a methodology, including an epistemology that was more indigenous to their own context, such as dialectical materialism (Chen & Chen, 2012; Slife & Melling, 2012). These Eastern researchers saw quite clearly the Western philosophies that were endemic to “the” scientific method being taught in the West and appropriately desired to explore alternatives. Too often, however, this clear identification of Western method biases is obscured with claims of value-free objectivity.
What, then, are these assumptions of psychology’s empiricism and how might the potential for these kinds of prejudices practically affect the study of certain psychological phenomena? Perhaps the most obvious “prejudice” in this regard is the simple empirical injunction that “only the observable can be properly known.” As mentioned, this prejudice literally means that only that which comes through our eyes can be known and/or measured. This meaning is a prejudice because it is an unproven, value-laden judgment about what has worth in science – the observable has worth. Moreover, this prejudice ignores the considerable practical evidence that humans have knowledge of many other forms, from their thoughts to their feelings to their relationships. Some empiricists might respond that these forms of knowledge are private, and thus not subject to public verification, which is surely true from this narrowed empiricist perspective. Still, this response begs the question of whether there are forms of nonobservable knowledge that are publicly verifiable. As mentioned above, the meanings of a book, whether storyline or information, do not “come through the eyes,” yet people can experience these meanings and come to similar conclusions about what books mean.
If, however, the psychologist persists in using empiricism in the narrowed sense, which is true even when operationalizations are used (see above), then only the observable portions of psychotherapies will be emphasized. This emphasis implies that some portions of psychotherapies will not be studied, the unobserved portions. An example involves what some would call the “healing relationship” between the therapist and client. As important as this relationship is (Norcross, 2002; Slife et al., 2005), the “betweenness” of this relationship is not strictly observable. The therapist and client, as bodies and behaviors, clearly “fall on our retinas,” but the interpersonal relationship between them does not.
This empirical situation also has important implications for what is considered “evidence-based practices” in psychology. Not only do important aspects of therapy remain unstudied but also those therapies that emphasize observables are more easily studied. Behaviorism, for example, stresses observables almost exclusively. Indeed, behavioral accounts of therapy are routinely understood to have inherently empiricist theoretical foundations (Rychlak, 1981), making these therapeutic strategies more connected with and amenable to empirical scientific methods. As a result, those therapies that are more conceptually related to empiricism are those typically approved as evidence-based practices (Messer, 2001, 2004).
Existential therapy, as a counterexample, will likely never become an empirically based practice because existentialists contend that vital elements of their therapy are not observable (Yalom, 1980). The therapist-client relationship is just one such element. If this contention is true, then existential therapy will be poorly investigated by empirical scientific methods and likely omitted as an evidence-based practice. Those who advocate empirical scientific methods may contend that the observable aspects of existential therapy are the more important aspects, but this contention is the method tail wagging the therapy dog. In other words, it is less about what existentialists consider existential therapy and more about what is in the service of the method.
If these psychotherapeutic implications have merit, then the ideology of empiricism has rather dramatic economic implications because certain therapies, those that agree more with empiricism’s prejudices, are more likely to be included on the list of evidence-based practices regardless of investigation. Therapies that are omitted from this list might be considered not only less effective but also ineligible for reimbursement from healthcare insurance companies. The point is that all these economic outcomes are driven not by the data of an objective world, but by an empiricist ideology.
The philosophy of empiricism is also known to accompany and perhaps even complement other ideologies. Although empiricism is obviously not value- or bias-free, given our previous discussion, the widely held notion that empirical evidence is objective or relatively bias-free evidence may stem from its association with other ideologies, such as logical positivism or even liberal individualism. Liberal individualism, for example, has been defined as the relatively unimpeded pursuit of freely chosen ends in the promotion of individual autonomy (Fowers & Richardson, 1993; Richardson, Fowers, & Guignon, 1999; Taylor, 1985). Often considered a political ideology, liberal individualism has been “conceived as a means to free individuals from arbitrary authority and oppressive bonds” (Fowers & Richardson, 1993, p. 355). Arbitrary authority, in this sense, is the imposition of unjustified values or biases, particularly on an individual.
As dissimilar as individualism and empiricism may at first seem – with the former a philosophy of politics and the latter a philosophy of science – the two ideologies have a similar distrust of arbitrary values and biases. The individualist resists the imposition of arbitrary values to protect individual autonomy and the empiricist resists the imposition of arbitrary values to protect the objectivity of knowledge. Arbitrary values are those considered merely personal or subjective so that when both ideologies resist these subjective values they can both be viewed as moving generally away from subjectivity and toward a more objective understanding of the world.
This somewhat complementary relationship between empiricism and individualism is also clarified in their shared “prejudice against prejudice.” The liberal individualist seeks to prevent arbitrary forms of moral authority to protect individual rights, etc., and is thus prejudiced against arbitrary forms of moral prejudice. Similarly, the empiricist seeks to prevent biases and subjectivities to protect more valid forms of evidence, such as sensory experiences, and is thus prejudiced against nonempirical forms of prejudice. Although the two ideologies do not logically necessitate one another, their seeming complementarity can lead them to be confounded in certain political or scientific arenas, such as the ethics of science (cf. Abou, 1995; Haan, 1982).
The general point here is that empiricism is not a conception or method for mapping an objective reality; it is an ideology for illuminating various aspects of an interpreted reality. That this reality is interpreted is not necessarily negative. It is only negative if one accepts the prejudice against prejudice and then overlooks that this acceptance is itself a prejudice. All methods and epistemologies, in this sense, are interpretations of reality. What is pivotal from this perspective is not only being aware of this interpretation but also taking it into account when considering method outcomes, especially power and economic relations.