Save your time - order a paper!
Get your paper written from scratch within the tight deadline. Our service is a reliable solution to all your troubles. Place an order on any task and we will take care of it. You won’t have to worry about the quality and deadlinesOrder Paper Now
Bruce A. Thyer
Becoming a theorist is not a mysterious process. It works in much the same way as becoming a researcher or teacher—through formal and informal learning experiences.
—Randolph-Seng (2008, p. 39)
eretofore, the chapters in this handbook have been concerned with dif- ferent research methods to systematically gather empirical data for vari- ous purposes. Sometimes the purposes of empirical research are exploratory in nature—to learn about some phenomenon of interest, to become more informed about it, or to use it as a preliminary for addi- tional work. On other occasions, we gather data for purely descriptive
purposes, to be able to quantify variables of interest or to learn more about the subjective experiences of people such as social work clients.Much social work research is explanatory in nature, intended to help us understand linkages and relationships among variables and sometimes to try and develop causal knowledge about the etiology of psychosocial prob- lems. And another useful type of research is the development and testing of methods of social work assessment such as rating scales, rapid assessment instruments, and indices. The least common but perhaps most valuable type of social work research are evaluation studies, actually looking at the processes and outcomes of social work services, programs, and policies. Whether the research method involves quantitative measurement or qualita- tive data or uses a mixed methods approach, thus far we have been discussing methods involving the gathering of empirical data, with such studies being generally lumped under the term of empirical research. Empirical research itself may be divided into two types, basic empirical research (which is, roughly speaking, aimed at developing an understanding of psychosocial phenomena) and applied empirical research (aimed at solving problems). Social work tends to involve very little basic empirical research, which is understand-
able given our profession’s focus on solving serious psychosocial programs, a focus almost synonymous with the term applied.We are a practice profession, not an academic disci- pline studying things for their own sake. The public expects us to develop effective solu- tions to problems, and we study theory primarily to advance that purpose. Turner (1979) expressed this view in the following statement:
The clinician’s principal interest is in the utility of theory: what can it tell me about this situation that will permit me to act effectively? It is, therefore, not knowledge for its own sake, but knowledge for use. (p. 3, italics added)
CHAPTER 26 � THEORETICAL RESEARCH 469
This is a pragmatic view similar to that described about two decades earlier:
Social work is not a science whose aim is to derive knowledge; it is a technology whose aim is to apply knowledge for the purpose of control. Therefore, on the research continuum social work research falls nearer to the applied end, because its purpose is practice knowledge. (Greenwood, 1957, p. 315)
A social worker in his capacity as a social worker is not a social scientist. His role as a practitioner is not primarily composed of observing, understanding, doing research, and developing theory. As a practitioner he must act, and act with what- ever tools and knowledge he has at his command. . . . The primary value of social science content for social work . . . is its relevance to practice. The ultimate test of such relevance is what difference the content can make in social work’s capacity to help people. (Stein, 1960, p. 200)
The present chapter is devoted to a topic different from either basic or applied empirical research, a field of scholarship labeled theoretical research, which, for the purposes of this discussion, refers to original explanatory formulations advanced without the benefit of systematic and originally collected empirical data. Randolph-Seng (2008, p. 39) provides a similar description: “A theorist examines a systematic collection of evidence to create ‘intellectual structures’ that make sense of diverse findings and, in turn, provide distinct predictions that can be examined empirically.” Table 26.1 presents some common definitions of theory derived from the social work
literature presented over the years, definitions that represent the sense in which the term is used in this chapter. Other conceptions can be presented as to what is meant by the word theory, but the usage of the term in this chapter is the most commonly understood and accepted meaning of the word. Social work scholarship focusing on theoretical research of any type is rare. Daley,
Peters, Taylor, Hanson, and Hills (2006) examined 885 articles retrieved from 30 social work journals published in 2002 and categorized them as to the type of research that was represented. More than 41% of the articles were empirical studies with no theoret- ical linkages, and 30.7% were articles without any empirical or theoretical foundations (e.g., about 71% of the articles contained no theory discussion at all). Few articles (9.5%) contained any content representing the advancement of theorizing. Lest the reader think that the year 2002 was an aberration, Gentle-Genitty et al. (2007) partially replicated Daley et al. (2006) by retrieving and examining 1,168 articles published dur- ing 2004 that were obtained from 37 social work journals. Only 6% (n = 71) met the cri- teria for contributing to theory development with an empirical base, and fully 12 of the journals (32%) did not contain any articles involving theory development and explana- tion. These authors concluded that “overall, the results indicate minimal focus on theory within the 37 journals selected for review” (Gentle-Genitty et al., 2007, pp. 69–70). Given the crucial importance that is ascribed to good theories within social work (Thyer, 2001a, 2001b, 2008c), the relative lack of attention to theoretical developments within our field is rather shocking.
Types of Theories
Theories themselves can be roughly categorized into those focusing on description, expla- nation, prediction (e.g., creating hypotheses), and prescription (e.g., interventive theory), with grand theories encompassing all four of these aspects and being applicable across a wide range of social work practice and psychosocial phenomena. Some examples of grand theories include psychoanalytic theory and its numerous psychodynamic offshoots; social learning theory, comprising operant, respondent, and observational learning principles; general systems theory and its social work variants, such as the ecosystems perspective; and so forth. Generally, grand theories provide conceptual frameworks to explain normal human development across the life span, the development of dysfunctional behavior across the life span, and behavior occurring at the level of individuals, families, small groups, organizations, communities, and society as a whole. Obviously, relatively few grand theories fulfill these ambitious goals, and those that do possess such wide applicability gen- erally do so in theory only, so to speak, not in empirically supported applications. In the fullness of time, it is hoped that the social and behavioral sciences will move toward a grand unified theory of human behavior embracing actions at all levels of analysis: individual, couples, families, small groups, organizations, communities, nations, and larger units of analysis. This would be an idealized goal analogous to the search for a grand unified theory (a.k.a. GUT) within the discipline of physics, one that mathematically unites the principles of gravity, strong and weak forces within the atom, and electromagnetism.Although we are
470 PART V � CONCEPTUAL RESEARCH
• “Theory consists of an interlocking set of hypotheses that are logically related, and it seeks to explain the
inter-relationships among empirical generalizations” (Tripodi, Fellin, & Meyer, 1969, p. 13).
• “Theory in social work is intended to provide explanations for the phenomena of practice” (Lewis, 1971,
• “For the clinician seeking to offer responsible, effective, intervention, the essential and important
contribution of theory is its ability to predict outcomes, or, in other words, its ability to explain” (Turner,
1979, p. 8).
• “Theories are sets of concepts and constructs that describe and explain natural phenomena” (Tolson,
Reid, & Garvin, 1994, p. 241).
• A “collection of related statements or propositions that attempt to describe, explain, or predict a
particular aspect of experience” (Queralt, 1996, p. 12).
• “A group of related hypotheses, concepts and constructs, based on facts and observations, that
attempts to explain a particular phenomenon.” (Barker, 2003, p. 434).
• “The essential idea of theory is that it is assumptions, that it can systematise, and that it can explain
phenomena” (Johnsson & Svensson, 2005, p. 420).
• “A systematic set of interrelated statements intended to explain some aspect of social life or enrich our
sense of how people conduct and find meaning in their daily lives” (Rubin & Babbie, 2008, p. 646).
• “A reasoned set of propositions, derived from and supported by established data, which serves to
explain a group of phenomena” (Grinnell & Unrau, 2008, p. 558).
• “A more or less well argued explanation of reality. Collectively, then, theories are explanations of
reality—of human behaviour or particular social phenomena” (Gray & Webb, 2009, p. 10).
NOTE: Italics added in all definitions.
TABLE 26.1 Some Selected Definitions of the Term Theory
CHAPTER 26 � THEORETICAL RESEARCH 471
obviously a very long way from having such a theory of everything in social work, there is a long tradition of sober-minded thinkers who have explored the possibility of this within the social and behavioral sciences and even an eventual unification of the behavioral and physical sciences (see Causey, 1977). Others deny the theoretical possibility of having a theory of everything, but then heavier than air travel was once deemed impossible also. Absent satisfactory grand unified theories or less large-scale theories, we are often con-
tent to contend with a lesser level of explanation, called middle-range theories, which are more limited in scope and address a circumscribed area or domain of practice (see Martin & Sugarman, 2009). These might address a particular client group (e.g., Afrocentric theory or adolescent development), the etiology of a particular problem (e.g., the tension reduction theory of alcohol abuse, the stress-diathesis theory of schizophrenia), or the presumptive mechanisms of action explaining how a specific intervention is supposed to work (the theory behind the principles of Alcoholics Anonymous or the theory behind cognitive therapy for depression). Obviously, there are far more middle-range theories than there are grand theories in social work, but it would be hoped that in the fullness of time and the accumulation of data that middle-range theories would be legitimately inte- grated into one or more grand theories. This level of theory has been described by the sociologist Robert K. Merton as follows:
Merton asks sociologists to be more modest in their goals when developing theories. Instead of attempting to develop highly general, and therefore abstract, theories to explain the structure and functioning of total societies, he proposes a series of less general theories be developed to explain smaller components of social reality. Thus, Merton’s approachwould have sociologists concentrate on developing theories of deviance, strat- ification, minority-majority relations, population change, and so forth that are less abstract but more closely tied to empirical reality, thus more testable. Only after a number of such theories are developed should attempts bemade tomove up the level of abstraction and develop more general, all encompassing theories. (Chafetz, 1978, p. 6)
Lesser still in scope are so-called low-level or practice theories, theories arising to account for highly specific situations. While appealing in many ways—they can make intuitive sense, may require lesser levels of abstraction, and can be developed for most situations—low-level theories threaten to overwhelm the profession with a plethora of competing perspectives. Loeb (1960) referred to them in this way:
Low level, or ad hoc, theories are annoying. They fill the literature with research reports of little significance or relevance. On the other hand, a few of them have important effects in modifying middle-range theories. These pearls may be worth the task of opening all the oysters. (Loeb, 1960, p. 8)
Theoretical advances are very highly valued within the world of academics and often seen as a somewhat mystical process, particularly by practitioners. Creating a new theory is seen as more prestigious than extending an established one, and as a consequence, sometimes existing theory is simply reframed in new language to give the appearance of having done something more novel than may actually be the case. The distinguished psy- chologist Walter Mischel (2009) made these observations on the subject:
Our urban legends, and the reality that underlies them, directly reward research and theorizing that emphasizes novelty and newsworthiness in publishing, grant get- ting, and tenure requirements, especially in highly competitive universities. You get
high marks for autonomy and brand new theories, unconnected to anybody else’s ideas, especially those of one’s mentors. That’s great, when the contribution is really new and solid, and adds to building a cumulative science. It’s not great when it turns out to be mostly recycled—repackaged with a more newsworthy sexy title . . . psy- chologists tend to treat other people’s theories like toothbrushes; no self-respecting individual wants to use anyone else’s. (p. 3)
One highly respected approach to social work is called task-centered practice (TCP; see Reid, 1992). Unlike most other approaches, this model is one of the few that was concep- tualized from within the profession, by and for social workers, to be applied to social work practice. TCP enjoys a solid empirical foundation (see the November 2008 special issue of the journal Research on Social Work Practice, focusing on this one approach) and has gen- erated a modest literature in its own right. However, Gambrill (1994) critiqued TCP as being essentially (my words here) watered down behavior analysis and therapy, failing to sufficiently acknowledge the essential similarities between TCP and behavioral methods and understating the valuable role that social learning theory plays in explaining how cer- tain methods of TCP assessment and intervention may work. Similarly, Thyer and Myers (1997) reviewed the essential similarities between the person-in-environment (PIE) per- spective so widely accepted within social work and the person and environment interac- tionist learning theory foundations of behavioral methods (which considerably predated PIE in social work), noting,
The parallels between the two are striking, so much so that it would seem that con- temporary social workers are reinventing the wheel, woefully ignorant of the social learning theory antecedents of current models or are simply choosing to ignore the similarities between the two approaches. (p. 21)
The psychotherapy called eye movement desensitization and reprocessing (EMDR) has been similarly critiqued as a weak form of cognitive behavioral therapy, and the orig- inator of EMDR also has been frankly criticized as engaging in making exaggerated claims and hucksterism as a means of promoting this technique, at great personal financial profit. Attempting to read the minds of social work theorists is always a risky proposition, but the temptation to relabel existing theory and to pass it off as a new development, per- haps with new language and proposed causal mechanisms, is undoubtedly a real dynamic in the world of theoretical development. It adds conceptual confusion and violates the principle of parsimony discussed in Chapter 1 of this book.
What Are Theories Made of?
Philosophical and Conceptual Assumptions Theories can be said to comprise a variety of elements. At the highest level of abstraction, a theory can posit a general worldview or set of philosophical assumptions. Most theories in the social and behavioral sciences, including social work, implicitly or explicitly accept the philosophy of science called positivism, which holds “that social behavior could be studied and understood in a rational, scientific manner—in contrast to explanations based on religion or superstition” (Rubin & Babbie, 2008, p. 642). Although subjected to many differing interpretations and often misunderstood, the relatively benign definition of positivism provided above remains the mainstream perspective of social science (also
472 PART V � CONCEPTUAL RESEARCH
CHAPTER 26 � THEORETICAL RESEARCH 473
see Thyer, 2008a). A contrasting view is called social constructivism, which maintains that much of our reality is socially constructed. The two views are not antithetical, of course, but they can serve as polar opposites anchoring our initial attempt at describing what goes into a theory’s umbrella perspectives. It should be acknowledged that these funda- mental philosophical underpinnings of social work theory cannot themselves be sub- jected to scientific scrutiny. They are assumptions that various theories accept or reject, recognizing that these propositions cannot be proven to be correct (or incorrect). In this sense, one’s philosophical positions can be seen as similar to the beginning axioms with which Euclid developed his system of geometry. Accept these axioms, and a very elaborate theoretical construction becomes possible.
Propositions The next level of abstraction can be called a theory’s essential propositions, general state- ments about the empirical world that is the focus of theory. To illustrate, here are a few examples of some fundamental propositions derived from psychodynamic theory:
• A substantial part of the activity of the human mind consists of unconscious processes.
• Unconscious processes are involved in cognition, affect, and motivation. • Important motivators of human activity are one’s sexual and aggressive urges. • People can feel things without knowing they feel them, and they can act on feelings of which they are unaware.
These are just a few representative ones, by no means suggestive of the comprehensive picture of the richness of this conceptual framework (e.g., see Westen, 1998). Social learning theory has its own propositions, such as the following:
• Human behavior consists of what we do—both observable behavior and unobserv- able behavior. Overt acts, covert speech, thoughts and cognition, feelings, dreams: all those phenomena people engage in are considered behavior.
• To a large extent, much (but not all) human behavior is learned through life experi- ences. This learning occurs throughout the life span.
• It seems very likely that similar fundamental learning processes give rise to individ- ual behavior across cultures and life circumstances and accounts for both normative and many so-called dysfunctional actions, feelings, and thoughts.
• Interpersonal behavior is also a function of these learning processes, giving rise to dyadic, group, organizational, community, and societal phenomena. These larger scale activities are, to a great extent, a more complex operation of fundamental learning mechanisms. (Thyer & Myers, 1997, p. 19)
And cognitive theory has its own propositions as well, such as the following:
• Individuals respond to cognitive representations of environmental events rather than to the events per se.
• Learning is cognitively mediated. • At least some forms of cognition can be self-monitored. • At least some forms of cognition can be altered. • Altering cognition can change dysfunctional patterns of emotion and behavior. (Thyer & Myers, 1997, p. 29)
It is always desirable for a scientifically sound theory to clearly outline its foundational propositions, and these should be clearly explicable in everyday language. From such propositions can be derived specific, predictive hypotheses, statements about what can be logically concluded to happen under certain circumstances within everyday life using nat- uralistic research methods or in the deliberately contrived or manipulated situations we call experiments or demonstrations. Clinical evidence of the validity of the above psycho- dynamic propositions is seen by some theorists through phenomena such as slips of the tongue, neurotic symptoms, jokes, mistakes, and the recalled contents of dreams. And more systematic evidence may be sought through more structured research methods such as surveys, quasi-experiments, experiments, and longitudinal studies of the psychosexual development of infants, youth, and adults. The following are some actual hypotheses derived from psychodynamic theory and its foundational propositions.
Hypotheses 1. Panic attacks are caused by a buildup or discharge of unfulfilled libidinal urges (e.g., sex drive).
2. Individuals suffering from panic attacks do not have a healthy sex life.
3. Individuals with panic attacks who subsequently develop a healthy sex life, with regular orgasms, will experience a reduction in panic attacks.
(I am not implying that the above hypotheses are anything but representative specula- tions to be found in this theory’s literature.) To test these hypotheses, one must opera- tionally define the variables involved, and thus variables are an important aspect of any theoretical structure.
Variables Panic attacks can be defined in various ways, but the operational definition found in the text-revised fourth edition of the Diagnostic and Statistical Manual of Mental Disorders (American Psychiatric Association, 2000) is the most widely used approach to conceptu- alizing this phenomenon. Defining a healthy sex life is of course subject to widely varying interpretations, but such definitions would likely involve sexual activities that at least occasionally lead to orgasm.And the experience of an orgasm could be described in phys- iological and affective terms. With these operational definitions in mind, we could then try and test our hypotheses by gathering pertinent data, with observations and empirical evidence also being an important part of any theoretical structure.
Observations or Empirical Evidence Any good theory for social work should be supported by a credible body of empirical evi- dence, information systematically gathered about the world using both reliable and valid methods of assessment and capable of being replicated by others. Omit any of these adjec- tives—systematic, reliable, valid, or replicable—and you have strayed from the domain of science. Evidence can take many forms, including but not limited to clinical observations, correlational studies, historical analysis, naturalistic studies, epidemiological studies, single- subject studies, preexperiments, case control studies, quasi-experiments, time-series designs, randomized controlled trials, meta-analyses, and systematic reviews, among other types. Generally, the more good evidence bearing on a given theory, the better. The
474 PART V � CONCEPTUAL RESEARCH
CHAPTER 26 � THEORETICAL RESEARCH 475
more this evidence consistently points to a particular conclusion (e.g., the theory is cor- roborated, or the theory is refuted), even better still. We could attempt to test Hypothesis 1 listed above using a variety of methods. One
approach would be to obtain a sample of adult individuals who experience panic attacks and a demographically similar group who do not. We could ask them to retrospectively create a diary of sexual activities for, say, the past 2 weeks, specifically noting their orgas- mic experiences. If the panic-prone individuals reported fewer orgasms than the nonpan- ickers, we could assert that the hypothesis was corroborated or supported and that the evidence provides some tentative evidence of one of Freud’s propositions. Or, if no differ- ences were found, we would note that the hypothesis was not confirmed (or discon- firmed, if you prefer stronger language) and that a selected Freudian proposition thereby weakened (or falsified, if you prefer stronger language). But this is not the only way we could test Hypothesis 1. A perhaps stronger method would be to ask our sample of pan- ickers and nonpanickers to prospectively keep a sex diary of orgasmic experiences and could then examine the results to see if the reports differed between our two groups. In science, prospectively collected data are usually accorded greater weight as credible evi- dence than is retrospectively obtained information. A third way to test Hypothesis 1 would be to examine nonpanicking individuals who subsequently experienced prolonged periods of sexual abstinence (e.g., religious people who take vows of chastity, crewmem- bers on submarines), to see if the predicted development of panic attacks occurs. Hypothesis 3 might be tested by following a cohort of sexually abstinent panickers over
time, some of whom later naturally develop sexual relationships and some of whom do not, and see if the frequency or intensity of panic attacks diminishes in the former group. Perhaps the reader could conceive of a more experimental or deliberately manipulative way to test Hypothesis 3? Over time, perhaps decades, heuristic hypotheses derived from theoretical proposi-
tions get tested in various ways, experimental and nonexperimental, and the validity or invalidity of the theory itself correspondingly waxes and wanes. Ideally, incorrect theories are eventually discarded, replaced by more fruitful, valid ones, recognizing, of course, that any conclusions regarding a theory’s truthfulness are best regarded as provisional and tentative, always subject to be revised as new or better data accumulate (Thyer, 1997). A theory may be revised in light of new data but should not be so elastic in its propositions that it is impossible to falsify. Such a theory lacks one of the essential features of a gen- uinely scientific accounting and will ultimately prove a nonproductive path.
What Are the Characteristics of Good Theories?
The theories used in social work range from the very weak to the scientifically strong. There is some consensus that the stronger scientific theories possess most if not all of the attributes listed in Table 26.2. But it must be admitted that relatively few theories in social work can be said to completely adhere to these rigorous benchmarks. Take some time to review these standards, derived from a variety of social work and non–social work sources (e.g., Chafetz, 1978; Polansky, 1975; Thyer, 2008a). A number of rating scales have appeared in the social work literature whereby one is supposedly enabled to quantify or rank the extent to which a given theory is said to possess selected attributes (reviewed in Daley et al., 2006). Some even allow you to come up with an overall score and thus hierar- chically rank the theories used in social work. I do not see this as a particularly useful task and do not recommend the exercise to my readers. None of these scales have demonstrated
sufficient acceptable levels of interrater reliability to make their findings trustworthy. Most, in fact, ignore the issue altogether, rendering any such ratings as liable to subjective bias. Each of the features listed in Table 26.2 can thus be said to be points for one to con- sider in appraising a given theory, recognizing that all will be found wanting to some extent. Other considerations may also be relevant, but the list in Table 26.2 can be seen as minimally essential qualities.
476 PART V � CONCEPTUAL RESEARCH
The Theor y—
is determinist—A good theory of human behavior, like theories for all natural phenomena, seeks
explanations for individual and social behavior in terms of environmental and biological factors, without
recourse to metaphysical or supernatural explanations. Determinism does not assert that free will does
not exist; rather, it focuses our attention on factors capable of scientific investigation and potentially
amendable to prediction and control.
is comprehensive—A good theory for social work practice should be as comprehensive as possible. A
theory that is developed to account for an isolated clinical phenomenon (e.g., depression) is not as
valuable as another theory that yields a plausible theory of depression, schizophrenia, personality
disorders, anxiety disorders, and so on.
is rational—A good theory must be rational. Given certain assumptions, subsequent deductions must not
violate the rules of logic, generate tautological statements, or yield contradictory conclusions.
is heuristic—A heuristic theory is one that generates predictive hypotheses capable of being tested and
in general stimulates empirical research to examine these hypotheses. A theory that generates little
research is seen as a less valuable one than a more productive approach.
is parsimonious—A good theory is one that effectively explains a large group of phenomena using
as few general principles as possible. Whenever one can choose between two theoretical
explanations, parsimony suggests that the simpler of the available and adequate accounts is more
likely to be true. Derived from the philosophical principle called Occam’s razor (from the 14th-
century philosopher William of Occam), parsimony can be expressed as “what can be explained by
the assumption of fewer things is vainly explained by the assumption of more things” (see R.
Epstein, 1984, p. 122).
is operational—A good theory uses clear and concise definitions of terminology, concepts, and variables
comprising that theory. Its language should lend itself to reliable and valid measurement.
avoids reification—To reify is to attribute reality status to concepts and explanatory factors that as of yet
have no real evidence that they truly exist.
avoids circular reasoning—Circular reasoning occurs when one invokes as an explanation a variable
whose only evidence for existence is that which it is said to cause. To say that a person runs away
from dogs because he or she has a phobia of dogs, when the only existence for the phobia is the
running away, is an example of this error. Saying that the reason why someone drinks too much is
because he or she has alcoholism, when the only evidence for the alcoholism is excessive drinking,
is another example. It is more likely to occur when metaphorical terms are reified and then used to
TABLE 26.2 Some Features Characteristic of a “Good” Theory
CHAPTER 26 � THEORETICAL RESEARCH 477
Where Do Theories Come from?
There is a wide array of sources of inspiration that can go into creating theoretical advances, sources that are available to virtually all social workers and thus enable any one of us— academic, practitioner, or student alike—to make potentially valuable theoretical contribu- tions. There is a common view that significant theoretical advances are made primarily by singularly creative individuals blessed with a gift of genius that enables them to produce rel- atively complete conceptual frameworks almost fully formed, like the goddess Athena sprang forth from the forehead of Zeus. And those of us lacking this gift of perhaps divinely inspired creativity are fated to never be capable of crafting important theoretical advances. Those of us lacking a theoretical muse can take heart from learning about the actual process of theory development from well-known individuals responsible for genuine theoretical breakthroughs. In this regard, the French genius Louis Pasteur said a couple of pertinent statements, one being “Chance favors the prepared mind.”And the second, “Let me tell you the secret that has led me to my goal: My strength lies solely in my tenacity.” If not the French Pasteur, how about the Americans Thomas Edison—“Genius is one percent inspira- tion and ninety-nine percent perspiration”—and Margaret Mead—“I learned the value of hard work by working hard” and “The way to do fieldwork is never to come up for air until it is all over,” or the British Thomas Huxley—“Patience and tenacity are worth more than twice their weight of cleverness” (all quotes retrieved from www.brainyquote.com). Here then are two important secrets—learn as much as you can about your subject matter and then work very hard, perhaps for a long time.Without adequate preparation and effort, you are virtually guaranteed to never create a useful theory. With these qualities, you have a chance at least to do so. Randolph-Seng (2008) provides these observations:
is falsifiable—The hypotheses derived from a given theory must be capable of being tested and of being proved
false. This can weaken a theory, and enough diverse instances of such falsification eventually cause the theory
to be discarded as incorrect. If a hypothesis (e.g., a prediction derived from a theory) is tested and the results
come out as predicted, we typically do not claim that the theory has been proven to be true. Rather, we say
that the theory has survived a test and been supported, corroborated, or strengthened. Since the supported
findings of any given theory (X) are subject to being overturned or subsumed by a better, more comprehensive
theory (Y), social work theorists are usually cautious about making assertions that a given theory has been
proven to be true.
is practical—For a theory to be practical for social workers, it should have some direct applications to
practice or policy. A theory lacking such practical applications is inherently less valuable for the human
service professions than one that does provide such guidance.
is empirical—A good theory is one that (eventually) is buttressed by solid empirical evidence. In general,
a theory with more relevant empirical evidence is to be preferred over a theory with less levels of
supportive evidence. A theory that does not lend itself to empirical testing at all, within the human
services, is generally seen as having little value.
has clarity—A good theory is clearly understandable by reasonably educated professionals. It is generally
free of jargon, and those unique terms or neologisms legitimately required by the theory are well defined
NOTE: These characteristics of a “good” theory are not listed in any priority order. All are important.
To theorize about something, you need to know a lot about it. The more knowl- edgeable you become about an area. . . , the more likely you will be able to identify gaps in existing theory and be inspired by your informed intuition an imagination to construct/extend a theory to bridge them. In turn, these bridges must be strengthened by constructive criticism and accumulation of evidence. (p. 39)
OK, hard work and intellectual preparation are crucial. But what are some other sources of theoretical advancement? Perhaps the greatest source of information that leads to theoretical advances in the human services consists of clinical observations, the impres- sions gleaned, sometimes systematically but often not, by our experiences as practitioners interacting with our clients. We listen to their stories, their histories, their truths, and sometimes their lies, and we glean corroborative information from their family members, peers, colleagues, and coworkers. We observe one instance of something with one client, repeated instances of the same thing with the same clients, and/or repeated instances of the same thing across a number of clients. If, over time, patterns seem to emerge, we may develop one or more theoretical accounts of the how and why of the phenomena on ques- tion. As Newmark and Beels (1994) suggested,
In psychotherapy, and in family therapy in particular, theories are rarely tested or even challenged by any quasi-experimental methodology. Instead they are con- structed from observations of repetitive patterns in the therapist’s office. In this sense they are drawn from a type of natural history methodology, rather than from experimental science. (p. 13)
A specific example can be drawn from Sigmund Freud’s practice with clients suffering from anxiety neurosis, work that led him to the following clinical observation:
In the case of agoraphobia etc. we often find the recollection of an anxiety attack; and what the patient fears is the reoccurrence of such an attack under the special circumstances in which he believes he cannot escape. (Freud, 1962, p. 81)
This clinical observation, coupled with other information about his anxious patients, led to an etiological theory that anxiety attacks were defenses against unacceptable sexual urges, such as to act promiscuously or to masturbate, leading to the avoidance of public places characteristic of agoraphobia. Apart from these and similar clinical observations, there were little empirical data to corroborate or refute Freud’s views for many decades. Aware of this gap, while employed as a clinical social worker at a specialty anxiety disor- ders program, I conducted a small-scale survey study of 28 members of an agoraphobia self-help group I was advising and found that every one did report having a history of panic attacks, consistent with Freud’s observations (Thyer & Himle, 1985). Moreover, when asked to provide an estimated date for the first panic attack and the emergence of their agoraphobia, there was a 9-year lag, with panics preceding agoraphobia. Fully 79% endorsed the statement that their panics caused their agoraphobia (as opposed to their agoraphobia caused their panics or that their panics and agoraphobia were unrelated), all consistent with Freud’s clinical impressions. Thus, in terms of descriptive features, Freud’s clinical observations have been supported by this simple survey and similar studies. However, his idea that panic attacks are caused by undesired sexual urges has not been supported, thus illustrating the problematic nature of clinical observations—sometimes they provide accurate leads, but sometimes they do not, and it can be difficult to sort out truthful clinical impressions from the false ones.
478 PART V � CONCEPTUAL RESEARCH
CHAPTER 26 � THEORETICAL RESEARCH 479
Apart from clinical observations, one’s skilled observations of so-called normal persons, children, youth, and adults can provide rich grain for the millstones of theory. Jean Piaget, the Swiss developmental psychologist, derived much of his theories pertain- ing to cognitive development from observing his own children. While tremendously influential for many decades, Piaget’s theory of mind has not stood up well to empirical testing and is no longer widely considered to be a viable model for understanding cogni- tive development among children. Again, while clinical impressions and anecdotes alone are great starting places for theoretical advancements and conceptualizing problems and treatments, they are generally seen as inadequate methods to separate valid theories from erroneous ones.
Prior existing theory is of course a productive inspiration for subsequent theoretical developments. As Isaac Newton is said to have stated, “If I have seen further it is by stand- ing on the shoulders of giants.” Reinventing the wheel is never a good use of time, and if one is thoroughly conversant with prior theoretical formulations surrounding an issue, one’s contributions are more likely to represent an advance as opposed to an unknowing reiteration of conceptual discoveries previously made by others. However, not all episte- mological frameworks within social work research take such a benign view of the effects of being immersed in prior theory. Many perspectives drawn from the field of qualitative inquiry take quite the opposite stance, suggesting that one’s familiarity with prior theory can bias or distort one’s observations (Shaw & Compton, 2003). It has been suggested by some qualitative methodologists that researchers should not deliberately immerse them- selves in prior theory until they have completed their data collection, so as to avoid this problem. This perspective is, however, a minority one.
Serendipity, or the role of chance, can also play a role in stimulating theoretical devel- opments. The German chemist Kerkulé is said to have discovered the ring-like structure of the chemical benzene by having a daydream about snakes holding their tail in their mouths (an ancient mythological symbol). My own work as a clinician brought me into contact with clients who suffered from trichotillomania (TTM), or compulsive hair pulling. One evening, I was having drinks and dinner at the home of one of my doctoral students, Patrick Bordnick, who, serendipitously, had two cockatoos (large, parrot-like birds) as pets. As we consumed our adult beverages, we watched with amusement the antics of the birds walking about the room, preening themselves, climbing curtains, and so on. Pat began telling me about a behavioral problem called feather-picking disorder (FPD) that can occur in birds. We began laughing over the apparent similarities between trichotillomania in humans and feather-picking disorder in birds. What began as a bibu- lous conversation subsequently led to a legitimate investigation of the two literatures, TTM and FPD; the recruitment of an avian veterinarian coauthor; and an eventual con- ceptual linking, resulting in a serious theoretical paper published in a psychiatric journal suggesting that FPD has the potential to be an animal model of TTM in humans (Bordnick, Thyer, & Ritchie, 1994). Were it not for the chance of Pat and myself being together with the cockatoos and cocktails, this theoretical paper might never have been written. Serendipity—it is a wonderful thing. One very legitimate source of crafting theoretical research is to extrapolate, to take
some existing theory and to apply it to a new field. Social work has a very long tradition of doing this, to the point of being criticized in some circles for having so little indigenous theory to point to, that is, disciplinary theory developed solely by social workers for use by social workers. Rather, we have tended to borrow existing theory from related disciplines such as psychiatry, psychology, and sociology and then present how this theory has etio- logical, assessment, or practice applications within social work. Psychoanalytic theory is of course the prototypical example of this type of theoretical borrowing, which ended up
having tremendous influence within the field of social work (Eisler, 1942; Goldstein, 2002); learning theory is another (Fischer & Gochros, 1975). More contemporary theo- retical borrowings have come from chaos theory (Hudson, 2000), the quantum theory of physics (Weick, 1991), and innumerable others. One becomes acquainted, perhaps super- ficially, with a theoretical framework originating in some other discipline; becomes intrigued with its possible applications to social work’s etiological understanding of some psychosocial issue, assessment, or intervention; delves more deeply into the new theory; and, in due course, prepares a paper introducing this new theory to the social work liter- ature. If the approach is seen to have merit, others may pick up the topic and elaborate on it, and slowly, perhaps, a sizable literature will emerge extrapolating this conceptual framework to our discipline, often over decades. Social worker Mark Mattaini (1991) engaged in this type of theoretical research when
he applied some principles of operant psychology to the problem of crack cocaine abuse. Dr. Mattaini used operant principles, previously not applied to this practice problem, to help conceptualize how crack use is so highly addictive and to propose various methods, all based on operant principles (matching theory, analysis of antecedents and conse- quences surrounding crack use, discriminative stimuli, etc.), to prevent and treat crack use at the level of individual, agency, community, and policy initiatives. Similarly, social work faculty member Karola Dillenburger (2007) attempted to inte-
grate operant theory with the field of practice known as victimology, the study of why certain people become victims of crime. By looking at factors such as an individual’s learning history, prevailing contingencies of reinforcement and punishment, establishing operations, the transgenerational transmission of trauma, behavioral economics, and cul- tural contingencies, Dillenburger fleshed out a beginning theoretical approach to this important emerging field of study. My own background in learning theory has led me to prepare a number of articles and
chapters, attempting to apply (extrapolate) certain aspects of this approach to various problems within social work. One of the most interesting such projects I have engaged in was to apply social learning theory principles to try and understand the causes of racial prejudice (Arhin & Thyer, 2004). By examining the literatures on observational learning, operant learning, classical conditioning, and attitude formation, we attempted to develop a theoretical model of why and how racially prejudiced behavior and attitudes are acquired and maintained by individuals and to provide some leads on how racial preju- dice may be reduced or eliminated. In another example, one from the animal learning theory literature, I knew of a phe-
nomenon called punishment-induced aggression (PIA). PIA is observed when, for example, you have two animals (e.g., two rats or two pigeons) contained in a small cage, and one animal only is subjected to random, brief, but painful electric shocks. A common reaction is for the shocked animal to attack the innocent bystander animal contained in the cage with it. It seems as if the animal exposed to aversive stimuli “lashes out” at its partner, who of course has nothing to do with the first animal’s exposure to painful shock. I was also aware of the literature on the antecedents to child abuse, many of which are macro-scale variables (e.g., poverty), but one commonly reported micro-scale phenome- non is the frequent clinical observation, buttressed by larger scale systematic survey stud- ies, that adult child abusers frequently lash out against children after the adult has been exposed to some aversive event. It may be something the child does (enuresis, encopresis, crying, irritating questioning, leaving a toy for the adult to trip over, etc.) or events com- pletely unrelated to the child (an argument at work or with one’s spouse, receipt of bad news, pain, etc.), which adversely affect the adult. Here is how one authority on child abuse describes a frequent dynamic found in such families:
480 PART V � CONCEPTUAL RESEARCH
CHAPTER 26 � THEORETICAL RESEARCH 481
Abuse incidents initiate with some behavior on the part of the child which the parent . . . perceives as aversive. (Kadushin & Martin, 1981, p. 253)
Aversive events are at the root of angry aggression. . . . Regardless of whether the aversively stimulated person is aware of these expressive responses and memories at the time, these initial reactions produce a tendency to attack the perceived source of the unpleasantness. (Berkowitz, 1983, p. 112)
The apparently spontaneous eruption of some instances of child battering bears a striking similarity to PIA observed in animals. How one construes (theorizes) the etiology of child abuse has an important bearing on how one undertakes to treat or prevent it. If you see the abuser as having a character flaw or psychodynamically construed develop- mental injury, then proper treatment will consist of some sort of therapy for the abuser. But if abuse is seen as related to PIA, then effective treatment and prevention will be aimed at eliminating the aversive events the abuser is exposed to, such as by providing treatment for a child’s chronic enuresis or intervention to alleviate chronic job stressors, if these are ascertained to be events that have evoked abuse in the past. This theoretical linking of PIA and child abuse was not intended as a comprehensive approach to under- standing but as an admittedly small conceptual contribution that may have treatment applications (Thyer, 1987a). In any event, it seems undeniable that learning theory vari- ables are implicated in the problem of child abuse. In another example of extrapolation, my clinical work exposed me to large numbers of
persons with panic attacks and agoraphobia who displayed escape-like behaviors that seemed very much akin to superstition. Such people may be unable to venture out of the home without holding on to some object that provides a sense of security (e.g., sun- glasses, canes, umbrellas, large purses, small bottles of liquor, or charms). They may engage in elaborate breathing exercises, counting rituals, or soothing affirmations, or they may carry bottles of ice water for drinking or spray bottles containing cold water for a quick misting of the face. Now both panic attacks and acute anxiousness caused by ven- turing outside the home tend to subside, even if the client does nothing, providing he or she remains in the anxiety-evoking situation without trying to escape. If someone has a panic attack or becomes highly anxious and uses some sort of comforting ritual or activ- ity, he or she may well feel better after a few minutes and not unreasonably attribute this diminished apprehension to what he or she did (e.g., an affirmation or breathing exercise) as opposed to the natural subsidence of fear. Over time, these people can become quite dependent on these superstitious-like behaviors to function, believing that they are needed to reduce anxiety. This can inhibit a full recovery. Again drawing from the experimental and theoretical literature on learning theory, I
was aware of a phenomenon called superstitious behavior, demonstrated experimentally in animals by Skinner (1948) and also among children and adults. The basic preparation in the lab involves exposing an animal to random positive reinforcement (e.g., food pel- lets) as it meanders around a cage. Over time, the animals often (not always) come to dis- play some repetitive or stereotyped behavior that is actually unrelated to the receipt of reinforcement but nevertheless is engaged in quite reliably. I speculated (Thyer, 1986) that the “comfort” behaviors observed clinically among anxious clients might be akin to the phenomenon of superstitious behavior seen in animals and adults (e.g., bowler’s gyra- tions after releasing the ball, a gambler’s rituals prior to the throw of the dice). A favorable outcome (e.g., diminished fear for the anxious person, food for the rat) is attributed to
what one was doing just prior to being reinforced with food or the relief of anxiousness. In reality, however, there is no “real” relationship going on. This was an admittedly spec- ulative paper, but the model does provide some implications for practice, and that is what theoretical papers are intended to do.
Sharing your nascent ideas with colleagues, fellow professionals, friends, or family can be a good way to help clarify your thoughts and to gain feedback about the clarity of your ideas and any gaps that may have escaped your attention. This can be done via formal brainstorming sessions held specifically for this purpose or, less formally, say, over a cup of coffee.
Consult the empirical literature on creativity. Interestingly, there is a small but intrigu- ing body of literature on the creative process. I particularly recommend R. Epstein (1996a, 1996b) and his essay titled “How to Get a Great Idea” in the 1996b volume.
Expose yourself to creative people and ideas. It has been said that creativity is infectious. One of the best ways to catch the creativity bug is to attend conferences, preferably unfa- miliar ones. If you have not been to the annual conference of the Society for Social Work and Research, consider attending this high-quality meeting (see www.sswr.org). Seek out interdisciplinary meetings or those sponsored by other disciplines related to your areas of social work, such as the annual meetings of groups like the American Psychological Association, the American Sociological Association, the American Public Health Association, the American Evaluation Association, and the Campbell Collaboration (especially recommended; see www.campbellcollaboration.org). Attend presentations by world-class researchers and theoreticians, chat them up during receptions and other social events, participate in special interest groups, and so on. Very frequently, I come away from major conferences with new book ideas, a new article I want to write, newly made professional friends with whom I subsequently collaborate, a commitment for a new scholarly project, and so forth.
Doctoral students are a particularly valuable resource for theoretical inspiration. If you are an academic with the opportunity to serve on doctoral dissertation committees or even to serve as a student’s major professor, such experiences often prove immensely productive in terms of stimulating theoretical thinking on your part. You must remain intellectually agile and will almost be forced to keep abreast of new developments in important areas of policy, practice, and theory to productively participate in doctoral education. Offer to teach a new class, one different from the stale old standbys you’ve been doing for years and ideally one with lots of theoretical content. University faculty are often allowed to take courses for free, and these can be terrific ways to bring you into contact with new theoret- ical developments. If you are a social worker with access to a university, take an advanced class in psychology, sociology, economics, or political science or attend colloquia, grand rounds, and seminars in new areas. If you are a student, cultivate the leading intellectual lights of your program—take their classes, ask to do an independent study with him or her, submit a jointly authored conference paper or journal article, and in general take advan- tage of all the opportunities afforded by the modern-day university for creative growth above and beyond the required program of study for your degree. Above all, begin by writing. Theories usually begin in embryonic form and take time to
fully develop. If you wait until you have a thoroughly fleshed out theory, you may wait a very long time indeed until you begin to write about your original ideas. Write out your beginning proposition or assumptions, explanatory variables, proposed relationships, and possibly some predictive hypotheses and how they could be tested. Perhaps make an out- line. Save it, and return to it periodically, adding paragraphs as your thoughts develop. This writing may not (indeed will likely not) reflect a linear progression. No matter. Perhaps keep a notebook to jot down your theoretical ideas or at least a dedicated documents file
482 PART V � CONCEPTUAL RESEARCH
CHAPTER 26 � THEORETICAL RESEARCH 483
on your computer. When ready, share and discuss with others, as noted above. Use feed- back to revise, tweak, and discard. Keep in mind that the best is the enemy of the good. If your standard is perfection, you will be immobilized by impractically high standards. Think of the flaws in existing grand, middle-range, and low-level theories and take com- fort in knowing that if these can be floating around with all their problems and incom- pleteness, there is room in the world of scholarly publishing for your own contribution. While a bit florid, the advice of Pannapacker (2009) is pretty good:
If you do not get an insight down on paper, and possibly develop it while your excitement lasts, then you are squandering the rarest and most unpredictable of your human capabilities, the very moments when one seems touched by the hand of God. (Pannapacker, 2009, p. B5)
The term exegesis refers to critical commentary upon a text, and it is most often used when the commentary is provided on holy scriptures. I suspect that the largest proportion of theoretical writing to be found in the social work literature comprises this type of con- tribution, writing and commenting on one or more specific theories that are being used in social work, with the purpose of familiarizing the reader with that theory’s fundamen- tal principles and actual or potential applications to social work policy and practice. Social workers typically come into contact with theoretical exegesis via undergraduate courses in psychology, sociology, and so forth and via authored textbooks used in our human behavior in the social environment (HBSE) courses. This type of writing can be helpful when a large amount of material must be covered or a smaller amount taught about a large number of theoretical perspectives. Of course, the latter information is often superficial at best, misleading at worst. Another avenue of communicating information about a large number of theories is to expose the students to edited books containing one or more chapters about a given theory, each authored by a presumptive expert in that theory. Some examples of comprehensive textbooks of this type include Roberts and Nee (1970), Dorfman (1988), Turner (1996), Brandell (1997), and Thyer (2008a). These allow somewhat more exposure to a given theory than the typical authored HBSE textbook, but even here chapters devoted to a particular approach may be only 20 to 30 pages long. A third method is an authored social work text that is devoted to a single theoretical per- spective. This can be the most comprehensive approach to conveying theoretical infor- mation and is particularly good when the author is a legitimate, noted scholar with expertise in that theory. Some examples here include Strean (1978) and Gambrill (1987). Such textbooks are not common in our discipline because we have, as a discipline, com- paratively few social workers who possess a reputation as serious original scholars and genuine experts within particular theoretical fields. Another drawback of the widespread practice of theoretical exegesis is that students
may only develop a familiarity with a given theory via second- or thirdhand sources, not primary ones. Thus, their views are derived from someone else’s interpretations of a given theory, rather than what the originators of the theory themselves actually had to say. Sometimes this is helpful if the original writings are particularly dense or obtuse or not translated into English. Having a knowledgeable social work scholar who has made it his or her business to become thoroughly familiar with a given body of theoretical content and to transmute this large literature into a summary form more readily digestible by the
B.S.W. or M.S.W. student can be a very helpful undertaking indeed. But this is only helpful if the translating (literally or figuratively) is faithfully and accurately undertaken. I myself have taken great offense at the misrepresentations to be found in the social work literature of my own preferred orientation, social learning theory, and have devoted a fair amount of time in documenting some of the more common errors and in providing cor- rective information (Myers & Thyer, 1994; Thyer, 1991, 2005). I can only imagine the indignation of my psychodynamic colleagues at the similar mishmash of psychodynamic theory, which is often found in social work writings. I recall participating in one doctoral student’s oral preliminary examination, wherein
as a part of his writings he had prepared a critique of the philosophical position called logical positivism. I asked him if he could tell me who Rudolph Carnap (one of the founders of logical positivism) was. He could not. He had not heard of him. I then asked him if he could name any original source materials written by any of the founders of log- ical positivism that he read in preparation for writing his critique of logical positivism.He had read none, being content to study instead only the flimsy social work literature criti- cal of this approach while never really coming into contact with what logical positivism really is! This is neither sound scholarship nor adequate conceptual preparation for social workers seeking a solid grounding in a given theoretical framework. If you want to learn about psychoanalytic theory, then for heaven’s sake read what Freud has to say, not (at least initially), what Smith says Jones had to say about what Freud wrote. If you want to claim expertise in learning theory, then read Skinner’s original articles and books, not commentaries upon commentaries upon Skinner. Once having developed this founda- tion of primary source materials, then go read contemporary commentaries and reviews, not claiming real familiarity with a given theory until you have achieved a good grasp of its foundational documents.
Theoretical Criticism as Theoretical Research
There seems to be a sort of Hegelian process operating within the social work theoretical literature in that shortly after a novel conceptual approach is introduced to our field (the- sis?), articles appear criticizing this new approach (antithesis?). Sometimes the critiques take the form of contending that the presumed new theory has been inaccurately described or represented by the pioneering author. Other times, the response is to con- tend that, actually, the new theoretical borrowing is simply not applicable to the social work processes and phenomenon it was originally said to relate to. More rarely, the criti- cism invokes empirical data supposedly indicating the serious lack of evidence, if not evi- dence in complete juxtaposition to, the proposed new applications to social work. The new approach may be beaten down by the critiques, to reside unmourned and rarely cited within the pages of textbooks and journals. Or the new theory’s proponents may rally around and defend their approach, perhaps with vigor and perhaps overcoming the pre- sumptuous critics (synthesis?). It is hoped that this is done in the dignified manner char- acteristic of scientific discourse, but this is not always the case. Recall that the purpose of theory is to bring us closer to nature’s truths. If a newly proposed or extrapolated theory (call it Theory X) is seen by some as taking us further away from nature’s truths, then some individuals may be prompted to criticize Theory X because of its perceived own shortcomings (a pure critique of X alone) or because X conflicts with the tenets of Theory Y or could perhaps be subsumed under Theory Y (a critique of X and a simultaneous pro- motion of Y). This too is legitimate theoretical critique.
484 PART V � CONCEPTUAL RESEARCH
CHAPTER 26 � THEORETICAL RESEARCH 485
When I began my M.S.W. studies in the mid-1970s, the textbook we used in my human behavior theory class was calledHuman Behavior in the Social Environment: A Social Systems Approach (Anderson & Carter, 1974). This volume was the nemesis for a whole generation of students who, like myself, found its extrapolations of the science of general systems theory, ecology, and cybernetics (themselves very legitimate fields) to social work theory to be exceedingly hard to grasp and apply. I hated it. I felt I was stupid for being unable to ascertain its value to working with clients, thought the jargon dehumanizing, and suspected it was simply a Potemkin-like exercise to beef up social work’s pretenses at being scientifi- cally based.However, I was always secretly bothered by the possibility that everyone else “got it,” and I did not. This was my burden of shame, and I was reluctant to reveal it to others. Imagine my delight when, two decades later, one of the smartest and most respected theo- retical scholars in our field (I mean this sincerely), JeromeWakefield (1996a, 1996b), wrote two brilliant and scathingly critical essays titled “Does Social Work Need the Eco-systems Perspective? Part 1: Is the Perspective Clinically Useful?” and “Does Social Work Need the Eco-systems Perspective? Part 2: Does the Perspective Save SocialWork From Incoherence?” I read with interest and a sense of self-validationWakefield’s conclusions that
the perspective’s lack of explanatory power . . . is its undoing . . . the perspective cannot contribute to assessment, integration, or bias correction . . . a careful read- ing of ecosystems writings . . . readily reveals the clinical vacuity of the perspec- tive . . . the continued widespread use of the ecosystems perspective could have dire consequences for the social work profession. (Wakefield, 1996a, pp. 27, 28)
The perspective had drained social work of its person-centered orientation and commonsense vocabulary, obscured social work’s value-laden purpose, and made it seem that social work is an intellectually flaccid enterprise that depends on untested perspectives rather than supported theories. (Wakefield, 1996b, p. 208)
There was some gentle pushback (Gitterman, 1996) but nothing that Wakefield could not handle well (Wakefield, 1996c). I highly commend this series of articles as illustrative of how one can make theoretical contributions to the field via the process of legitimately criticizing existing approaches. In lieu of the ecosystems perspective,Wakefield could not resist promoting his own preferred alternative unifying framework for social work prac- tice, an approach known as distributive justice; thus, he was critiquing Theory X as well as proposing Theory Y as a preferred alternative. Thanks to Wakefield’s example, I was emboldened enough to engage in some catharsis of my own in a later critique of systems theory (Gallant & Thyer, 1999). Each of the major theories residing within social work has been subject to theoretical
criticism. Here are a few representative examples:
• Solution-focused therapy has been criticized by the feminists (Dermer, Hemesath, & Russell, 1998).
• Feminism has been criticized by Lay and Daley (2007). • Afrocentric theory has been criticized by Pellebon (2007). • Psychodynamic theory has been criticized by Deal (2007). • Ecological theory has been criticized by Rotabi (2007). • Behaviorists have been criticized by the cognitivists (I. Epstein, 1975;Mahoney, 1989). • Cognitivists have been criticized by the behaviorists (Thyer, 1992). • Postmodernism (Brigham, 1977) has been criticized by Meinert (1998), Atherton and Bolland (2002), and Noble (2004).
The list goes on and on, with each such contribution being authored by someone convinced that he or she is doing the profession a service by exposing the fallacies of Theory X and providing greater light upon the path in his or her promotion of Theory Y. And to be sure, this sometimes does happen, which is a very good thing indeed and truly represents an advance in our theoretical understanding. Other times, to be sure, the effort is little more than shoving knowledge around with little in the way of theoretical progress. Nevertheless, theoretical criticism remains a solid way of contributing to social work’s
Another Example of Theoretical Research
Below is a description of another research project I was engaged in that led to a couple of original theoretical speculations, speculations subsequently corroborated by independent investigators through their own data-based studies.
Question—How Does Exposure Therapy for Phobias Work? In the early part of my career as a clinical social worker, I was employed in a specialty clinic that focused on providing services to clients with a primary anxiety disorder. Our interdis- ciplinary Anxiety Disorders Program was housed in a research-oriented Department of Psychiatry at the University of Michigan School of Medicine, and clinical and research roles were filled by psychiatrists, social workers, psychologists, nurses, and so on. By the 1970s, it was pretty well established that treatments involving exposing clients to fear- evoking situations under therapist-guided, carefully controlled, graduated conditions of suf- ficient duration were usually very effective in alleviating anxiety symptomatology associated with conditions such as specific and social phobias, panic disorder, agoraphobia, obsessive- compulsive disorder, and posttraumatic stress disorder (Thyer, 1983, 1987b), a conclusion that remains upheld today (see Barlow, Allen, & Basden, 2007). What was not at all clear was how, exactly, did exposing fearful clients for prolonged periods of time to frightening events result in symptom relief? In other words, what were the mechanisms of action of exposure therapy? There were a number of candidates, such as the psychoanalytic theory of abreaction, various learning theories (habituation, desensitization, reciprocal inhibi- tion, etc.), the so-called exposure effect written about by social psychologists, some emerg- ing cognitive explanations, and various biologically based accounts. The University of Michigan at the time was a hub for research into the newly discov-
ered compounds called endogenous opiates (EOPs), substances produced by the human brain under conditions of pain or great stress, and it had been speculated that EOPs may reduce fear and attenuate memories of fearful events. I had been involved in some prior research examining how the body responds to phobic anxiety in terms of endocrine and cardiovascular reactions (Nesse et al., 1985), and I used my familiarity with the EOP and endocrine release literature to speculate that phobic anxiety may also result in EOP stim- ulation. A physician colleague and I designed a simple A-B-A single-subject study with a voluntary client who had a severe phobia of snakes. The physician inserted a small needle into the client’s arm while she rested comfortably in a chair for about 60 minutes, during which time two blood samples were taken. Then I brought into the room, with the client’s informed consent, a large caged snake. I conducted exposure therapy with her for 30 min- utes, attempting to keep her at the maximum level of anxiety she could tolerate. During this time, two more blood samples were taken. When the snake was removed, she again
486 PART V � CONCEPTUAL RESEARCH
CHAPTER 26 � THEORETICAL RESEARCH 487
rested in the chair for 60 minutes, and two final blood samples were taken.We found that relative to the first baseline or A phase, during exposure therapy, her subjective anxiety became very high (self-rated at 90 points out of 100 maximum), her heart rate approached 150 beats per minute, and her levels of plasma beta-endorphin (a major vari- ant of endogenous opiates produced by humans) also spiked.We concluded our report of this empirical finding (basic empirical research?) by speculating that “one possible mech- anism for the therapeutic efficacy of prolonged exposure to anxiety-evoking stimuli is endorphenergic in nature” (Thyer & Mathews, 1986, p. 240).We then ventured in a small way into the realm of theoretical research by the following risky speculation:
One test of this hypothesis would be to conduct a control group study of simple phobics receiving a standardized period of exposure therapy, half of whom would receive infusions of a narcotic antagonist during treatment while the other half would receive active placebo infusions. If this infusion of a narcotic antagonist par- tially blocked the therapeutic response to exposure, EOP mechanisms would be implicated as a biological substrate for the efficacy of such exposure. Prior to con- ducting such an experiment, replication of the present study with a group design is certainly indicated. (Thyer & Matthews, 1986, p. 240)
Narcotic antagonists are sometimes used in the treatment of opiate abusers as they block the effects of drugs such as morphine or heroin. If clients who were pretreated with such compounds experienced reduced benefits from exposure therapy compared to clients not so pretreated, this could be seen as a robust test of the EOP hypothesis as an underlying biological mechanism at least partially related to the mechanism of action of exposure therapy. Subsequent empirical research has corroborated both of the hypotheses we put forth.
Phobic anxiety does result in EOP activation (Carr, 1996; Egan, Carr, Hunt, & Adamson, 1988), and pre-treating clients with an opiate blocker prior to exposure therapy does inhibit treatment response (Merluzzi, Taylor, Boltwood, & Gotestam, 1991). In the words of Kozak et al. (2007), “Results provide support for the endogenous opioid system as a potential underlying biological mechanism associated with behavioral changes during in vivo exposure” (p. 142). Now the issue is not settled, of course, but the above chronology extending over 20 years does illustrate how theoretical speculations can come to guide subsequent empirical research, which does, in turn, lead to further theory-based hypothe- ses. In this example, the original paper was empirical, but we concluded with a bit of the- orizing, and it was this small proportion of the paper that constituted our theoretical contribution. This speculation, originating in a small-scale, indeed a single-subject, study (which is about as small as you can go), led to further empirical testing and corroboration by others, with the new data in turn feeding back upon theory.
Theoretical research is a rarely undertaken form of scholarship, but an adequate theoret- ical foundation for understanding the causes of psychosocial problems and the mecha- nisms of action of social work interventions will ultimately be an essential feature of a mature social work profession. We are a long way from this, and such empirically sup- ported theoretical accounts will likely be developed using middle-range and lower range theories prior to the emergence of any comprehensive theoretical foundation covering the
large and diverse scope of social work practice. It is essential that we continue to borrow the best applicable conceptual frameworks from related disciplines and also develop a stronger indigenous body of theory emerging from those talented social workers most in contact with the world of practice. There are a variety of ways in which we can contribute to the body of social work theory, including exegesis (bringing relevant theory to the attention of the field), extrapolation (commenting on how heretofore unfamiliar theoret- ical accounts may be applied to selected social work phenomena), and theoretical criti- cism. We remain most in need of new original conceptual explanations that possess the attributes of a good scientific theory; that lend themselves to testing, falsification, or cor- roboration; and whose results can be fed back into the existing body of theory in a recur- sive manner that results in an improved theory closer to nature’s truth.We need always to keep in mind the epitaph found on the tomb of Karl Marx, one of the most influential social theorists of the 19th and 20th centuries, whose ideas continue to influence contem- porary political thought—“The philosophers have only interpreted the world in various ways—the point however is to change it.” The meaning and usefulness of social work theory reside in its consequences in action, as well as in fidelity to reality. A theory that does not ultimately produce helpful methods of alleviating human distress is a waste of time to the field. We have many such useless theories, and some are actively harmful (Thyer, 1994a, 1994b, 2008b).