.What Is a Theory?
Generally speaking, theories are abstract and general ideas that are
subject to rules of organization (see Box 1.1). These qualities may
be why the study of theory is often difficult for students. All of us are
more likely to be interested in concrete experiences, our own or those
of others. Theories are not the facts or data of experience, but they are
connected to those facts or data.
One complication in linking ideas and data is that the very same
data can be organized in different ways. As a result, different theories
can be used to make sense of the same set of facts. For example, how
are we to interpret the fact of a relatively high divorce rate? One
answer might be that the family as an institution is being threatened,
perhaps even disappearing. A different answer is that divorce weeds
out less viable marriages, thereby preserving the family institution. To
evaluate the merits of the two arguments, one must see what other relevant
facts are operating and how the entire argument is organized
into statements at a more abstract level.
One useful way to understand theoretical arguments is to examine
their basic components. We start with the idea that a theory is a
systematic collection of concepts and relations. This is consistent with
all four of the quotes in Box 1.1. Because there is an endless variety of
ways to organize ideas, we will limit ourselves to the requirements for
a scientific theory. Such theories contain systematically related propositions
that are empirically testable (Rudner, 1966). Thus, there are
several components that we need to discuss in further detail: (a) concepts,
(b) relations between concepts, (c) propositions, (d) relations
between propositions, and (e) connections between propositions and
the empirical world of observation. Before we look at each of these
components, we must distinguish between ideas and data.
Box 1.1 What Is a Theory?
“A scientific theory might be likened to a complex spatial network:
Its terms are represented by the knots, while the threads connecting
the latter correspond, in part, to the definitions and, in part, to
the fundamental and derivative hypotheses included in the theory.
The whole system floats, as it were, above the plane of observation
and is anchored to it by rules of interpretation. These might be
viewed as strings which are not part of the network but link certain
points of the latter with specific places in the plane of observation.
By virtue of those interpretive connections, the network can function
as a scientific theoryffrom certain observational data, we may
ascend, via an interpretive string, to some point in the theoretical
network, thence proceed, via definitions and hypotheses, to other
points, from which another interpretive string permits a descent to
the plane of observation.” (Hempel, 1952, p. 36)
“A theory is nothing—it is not a theory—unless it is an explanation.
One may define properties and categories, and one still
has no theory. One may state that there are relations between
the properties, and one still has no theory. One may state that a
change in one property will produce a definite change in another
property, and one still has no theory. Not until one has properties,
and propositions stating the relations between them, and
the propositions form a deductive system—not until one has all
three does one have a theory. Most of our arguments about
theory would fall to the ground, if we first asked whether we
had a theory to argue about.” (Homans, 1964, p. 812)
“Deductive theory can be described briefly as an attempt to
increase human understanding by providing explanations of
why certain things occur. It provides this explanation by having
a set of propositions and then deducing that, if these propositions
are true, and if certain other conditions are met, certain
specific and observable events occur. The more specific events
are then ‘explained’ by the more general propositions that have
been used as premises in deducing that the specific events occur.
The explanation is only as valid as the propositions and logic
that are used in the deduction, but one of the goals of science
is to gradually eliminate invalid propositions and increase the
number of useful, valid ones.” (Burr, 1973, p. 3)
“Theorizing is the process of systematically formulating and
organizing ideas to understand a particular phenomenon. A theory
is the set of interconnected ideas that emerge from this process.”
(Doherty, Boss, LaRossa, Schumm, & Steinmetz, 1993, p. 20)
Ideas and Data
Science is fundamentally concerned with ideas, data, and relationships
between them. Theory exists in the realm of ideas. Research takes
place in the realm of data. A science advances to the extent that its
theories and its studies (i.e., its empirical research) are productive and
mutually reinforcing. For knowledge to be scientific, scientists must
explain empirical observations by ideas or theory. Theories explain
by treating particular facts or observations as examples of general
principles or processes. For example, there may be certain conditions,
such as first intercourse, that encourages dating couples to become
more serious and less casual about their relationship. This would constitute
a hypothesis or testable conceptual proposition if data have not
yet been examined in this regard. If, however, we have a great deal
of data suggesting that first intercourse has a specific effect on couples’
“probability of marriage,” then that generalization is what we often
regard as a scientific fact.
The linkages between ideas and data can be organized in different
ways. If research produces empirical generalizations, the data might
inductively lead to the development of a new theory, or the data might
be interpreted in terms of an existing theory. When the process goes
from data to ideas, we think of it as being inductive. Sometimes, a
researcher starts with an existing theory, derives expectations about
the data (i.e., hypotheses), and then makes observations to see how
well the data fit. When the process goes from ideas to data, we think
of it as being deductiVe. Theoretical ideas in science are developed both
inductively and deductively. In either case, the ideas and the data must
fit together in a meaningful way for the theory to be judged adequate.
We caution against too rigid a distinction between ideas and data.
Sometimes, ideas can themselves be data. If the dating persons in our
example have ideas or beliefs about what might propel their relationship
to marriage, these ideas might be measured and become part of the
data. The ideas that people have may be beliefs, attitudes, or preferences.
These ideas might even be theories or explanations that people
have about such things as why couples have children at certain times in
a marriage. Thus, the ideas held by the people we study are often part
of our data. Even more important is the fact that our measures and
observations are founded on theories about the world and how we
can know and measure the world. So in many ways, scientific data are
constructed from ideas. We will have more to say on this point shortly.
This book is mostly about theoretical ideas that supply plausible
explanations about families. Over the years, family scholars have
offered many such explanatory ideas. We cannot call attention to or
restate every single such idea. Instead, we focus on the clusters of ideas,
or frameworks, that seem to be the most significant and popular now
at the beginning of the 21st century. New theories about the family will
be developed in the future. Some will build on one or more of the
currently existing theories, but others may have features that cannot
now be anticipated.
The question “From whom or where do theories and propositions
originate?” is important because it leads to a basic distinction regarding
theory. This distinction is between the “context of discovery” and
the “context of justification” (Kaplan, 1964). The answer to the question
about the origins of theory is simple: Theories and propositions
can come from anyone and anywhere. We might find a possible theory
in the writings of ancient scribes, a newspaper cartoon, or the observational
work of a research scientist. There should not be any restrictions
on the creative insight that might lead to fruitful theory building.
This is the “context of discovery,” and creative insight and thought
are its hallmarks. Grounding in the research literature and an understanding
of logic might be helpful, but they are not necessary conditions
for the development of theory. As a corollary, it follows that we
should not judge the usefulness or fruitfulness of a theory by its origin.
For example, if there are two proposed theories to explain ” adolescent
bed-wetting,” one from an esteemed research scientist in the area
and the other from your grandmother; we cannot decide on which is
better based on the status of the originator. Indeed, it is a hallmark of
the context of discovery in science that we do not judge theoretical
propositions by whether they originate with a political leader, a
grandmother, or a respected scientist.
The context of justification, on the other hand, is the context in
which theoretical propositions are put to the test. Here, we see if the
theory is justified in its claims. In science, this is a twofold process.
The first criterion applied is logical coherence. We ask, “Is this set of
theoretical propositions logically consistent with itself, and then is it
logically consistent with other well established theories in the area?”
The second criterion is the empirical adequacy of the theory. At this
point, we ask, “Do the deductions from the theory fit with our empirical
_.measures of phenomenon?” It is this step that many people
identify with “science.” However, during the evolution of a discipline
or area of study, it is obvious that the production of ideas (context of
discovery) must precede the testing of those ideas (context of justification).
Therefore, it is not unusual to find multiple approaches and
both qualitative and quantitative research strategies in areas that are
still developing basic theory.
This book will assist you in estimating the current status of
theoretical ideas about families. In the remainder of this chapter, we
cover the main ingredients that go into any scientific theory and then
suggest what makes theories about the family unique among scientific
theories. The next seven chapters separately cover the principal theoretical
frameworks that are important in family science today. Finally,
in the concluding chapter, we compare the major frameworks and
look forward to the future of theory in family science.
PHILOSOPHIES OF SCIENCE
The term philo s ophy can refer both to thevalue`s -and ‘-viray ,of
thinking of a person and to an academic discipline. Both meanings are
important in understanding the emergence and growth of family studies,
but for the moment we focus on the importance and relevance of
academic philosophy, especially the academic philosophy of science,
for scientific theories.
The methods of philosophy are logic and other forms of discursive
reasoning. Philosophers do not use the empirical methods of
science to establish the credibility of their ideas. Some philosophers,
known as philosophers of science, examine the principles by which
scientists work through reconstructing the logic of scientific processes.
Whereas sociologists of science systematically study the behaviors and
thoughts of scientists and the social organization of the sciences,
philosophers of science explore what science should or could he like
in terms of abstract ideas. There are several ways in which this philosophical
study is important.
The discipline of philosophy was historically,prionto the scientific
disciplines. The ancient Greek philosophers, among others, had many
creative ideas about human affairs and the workings of the natural
world. It was out of the-Enlightenment philosophy of the 18th and
19th centuries that the modern sciences in the Western world were
-born. During this period, it was common for individual scholars to
be both philosopher and scientist at the same time or to easily shift
back and forth between the two. It has only been during the 19th and
20th centuries that the various scientific disciplines have branched
into special fields to which a person could devote an entire career.
The historical legacy of philosophy as a precedent to the sciences
has meant that philosophy, as a formal discipline, has influenced all
sciences. Many of the influences have been indirect. The so-called
physical sciences, such as physics and chemistry, emerged first.
Between 1850 and 1900, the philosophical principles embedded in
these earlier sciences spilled over into the biological sciences, such as
botany and zoology, and the social sciences, such as psychology and
sociology. During the 20th century, as family science emerged as a
subspecialty within several of the social sciences, the philosophical
underpinnings continued to operate. Now, near the start of the
21st century, we are able to think of family science as a distinct field
with an interdisciplinary character or even as a unique discipline in its
In philosophy as in other academic disciplines, new ideas are
constantly being proposed and debated. A particular philosophy of
science may be popular at a given moment, but later it may be
challenged and some other philosophy may take its place at or near the
top – of the heap. Scientists, including family scientists, frequently turn
to the literature produced by philosophers to see what the current
issues are. When scientists need to justify the basic principles guiding
their search for answers to important questions, they sometimes seek
guidance from philosophers of science or from other philosophers.
It is now often asserted that there are different, if not rival,
philosophies of science. This is apparent in professional philosophy
itself and, more important for this book, among family scientists as
well. The philosophies we adopt influence the way we conduct our
scientific practice, including the theories we create and our attitudes
toward the theories of others. We sketch below a few of the central
philosophical views that may help to distinguish family theorists and
other family scholars from one another.
Key elements of three philosophies of science appear in Box 1.2.
We have called these positivistic, interpretive, and critical philosophies
of science (e.g., Neuman, 1994). We do not claim that these are
the only three alternatives in family science. Indeed, we discuss other
alternatives, such as feminism and postmodernism, in subsequent
chapters. At this juncture, we present Box 1.2 to illustrate the nature
of philosophical assertions. To further illuminate the issues involved,
we focus now on truth claims and values.
A group of philosophers working in Vienna in the early part of the
20th century argued that knowledge claims are either true or false and
that the job of scientists is to verify the true claims. Another philosopher,
Sir Karl Popper, disputed this view. Popper (1959) argued that
knowledge claims could be shown to be false but that there is no way
to prove them to be true. Popper’s position has become influential
among not only philosophers but also practicing social scientists.
Whether or not family scholars have read Popper, his ideas have
indirectly influenced family theorizing, helping to usher in what is
sometimes called a postpositive era in the philosophy of family science
(Thomas & Wilcox, 1987).
POSITIVISTIC View of Knowledge.
There are objective truths,
processes, or realities to be discovered about families.
Values: Family science can and should be value neutral if not value free.
Criteria for Evaluating Family Theories: Good theories should be
rationally constructed (e.g., internally consistent, simple, coherent, clear,
explicit, general, abstract). Good theories also should be empirically
relevant (e.g., testable, fit well with data).
Goals: Explanation and prediction.
Scholarly Style: Analytical, causal, deductive or inductive, deterministic
or probabilistic, factual, logical, materialistic, mathematical, mechanistic,
observant, planful, precise, quantitative, structural, etc.
INTERPRETIVE View of Knowledge. Truth is subjective, and
all knowledge about families is created by interpreting actors
engaged in conversations with one another.
Values: Family science is value relevant, and family scientists should
become aware of and open about their own values.
Criteria for Evaluating Family Theories: Good theories should have
literary qualities (e.g., elegance, imagination, narrative power). Good
theories also should be based on data grounded in the experiences of
Scholarly Style: Artistic, evocative, existential, hermeneutic, humanistic,
intuitive, metaphorical, phenomenological, postmodern, processual,
self-reflective, sensitive, speculative, spontaneous, symbolic, etc.
CRITICAL View of Knowledge. Truth is imposed by those
with the power to shape knowledge.
Values: Family theories are value laden. All values should be exposed
and challenged to create opportunities for change.
8 FAMILY THEORIES
Box 1.2 (continued)
Criteria for Evaluating Family Theories: Good theories contextualize
phenomena and allow for pluralism. Good theories also are emancipatory,
prescribe changes, display the theorist’s ethical stance, and fit well with
the theorist’s personal experiences.
Goals: Emancipation or empowerment of oppressed people and social
Scholarly Style: Constructivistic, dialectical, feminist, liberal or radical,
macroscopic, pluralistic, postmodern, processual, relativistic, etc.
A central difference between positivism and other philosophies of
science is their perspective on values. It should be obvious that no
human behavior is valueless in an absolute sense. Indeed, all of the
philosophical arguments we are considering are themselves evaluative.
To say that science should be without values is to value “valuelessness.”
To value something is simply to hold and to express a conception
of a desirable state of affairs.
One of the key reasons that science emerged historically as a way
of thinking and working that is somewhat different from the arts, religion,
and politics was concern about the distortions in knowledge
along with considerable human suffering that seems to result when
reasonableness, fairness, and facts are devalued. Science is not without
its faults, just as no individual human being or social organization
is faultless. The “value neutrality” sometimes advocated by scientific
positivists is similar in at least some ways to value tolerance or respect
for value diversity. But this is a matter of degree, and the valuerelevant
and value-laden positions push further the ideals of tolerance
and diversity, sometimes to the extreme point of complete relativism.
Writing several decades ago about science as a vocation, sociologist
Max Weber made some interesting comments:
Today one usually speaks of science as free of presuppositions. Is
there such a thing? All scientific work presupposes that the rules of
logic and method are valid. Science further presupposes that what is
yielded by scientific work is important in the sense that it is worth
being known. In this, obviously, are contained all our problems. For
this presupposition cannot be proved by scientific means. It can only
be interpreted with reference to its ultimate meaning, which we must
reject or accept according to our ultimate position toward life.
(quoted in Gerth & Mills, 1946, p. 143)
More recently, Christensen (1964) provided an analysis of the value
issue in family science. His assessment was quite Weberian.
Christensen advocated neither the rejection nor espousal of nonscientific
values by theorists and other scientists. Rather, he advocated
the identification and separation of nonscientific values from scientific
values. This may continue to be the most common view among family
scientists, but it is not the only view. According to the critical philosophy
in Box 1.2, for example, scientific facts are inseparable from the
values that scientists have about their subject matter.
What we have just said about values applies as well to ideologies.
The notion of ideology has three major meanings: (a) a set of beliefs;
(b) the systematic study of a set of beliefs, their nature, and their
origin; and (c) visionary speculation, often about ideals and with an
action agenda for achieving the ideals. There is no question that
systems of belief about families can be part of a family theory or that
carefully studying those beliefs might improve a family theory. There
is also no doubt that beliefs about family science can affect family
theories. Even visionary speculation has creative potential for
theories. Our caution is only that ideologies among family scholars
that are philosophical or political may be usefully connected with a
particular family theory, but they are not within or constitutive of
the theory itself. So if a scientific theory argues that X causes Y, the
truth of this does not depend on whether we like or dislike X or Y.
The theory may suggest how to change Y if the direction of change
is consistent with our values and ideology, but scientific theory
should be equally useful for someone with exactly the opposite values
To summarize, philosophical ideas establish principles that help
frame the ways that family theories and other aspects of scholarship
are created and used. Philosophical ideas are themselves foundations
for scholarship, but they cannot be scientifically proven true or false.
In many ways, scholars either consciously or unconsciously adopt
a particular philosophical stance in regard to research by the way
they do their research. Those who attempt not to use their research as
a vehicle for their own values, the values of their religion, or their
political values would be clearly identified as “positivist.” Although
those holding interpretive and critical philosophies have criticized
this perspective, the positivist approach remains the hallmark of
scientific work and scholarly accountability. Indeed, the public
expects family scientists to present their research findings independent
from their religious and political beliefs. In science it is expected
that any research finding can be replicated by other researchers
regardless of whether they have the same or different political and
religious beliefs. This is as close as we may be able to get to scientific
PARTS OF A THEORY
All scientific theories use the same basic building blocks. These
building blocks are concepts, relations, and propositions. The section
below discusses each of these central ideas.
Concepts are abstractions. A concept is not the thing but stands
for the abstract class of things, ideas, or entities. Concepts are essential
to theories because they enable us to organize experience. We do not
invent a new concept every time we refer to a unique event. For
example, we do not need a new term for marriage or a new meaning
for the event called marriage every time a wedding takes place. Instead,
we think of getting married in a more abstract way. We say that all
weddings have certain defining properties. If we want to distinguish
different kinds of weddings, we might establish categories. We could
distinguish elopements from formal weddings and church weddings
from civil wedding ceremonies. The categories remain abstractions,
however, in that they all refer to occasions of a particular type, not to
specific instances of weddings.
Social actors construct concepts for the purpose of communicating
meaning, whether the social actors are family members or family
scientists. So all concepts are “constructs.” However, we do not
whimsically invent concepts but rely on the common stock of terms
and definitions that already exist in common usage. Sometimes we
alter meanings or the words used for important ideas. Even dictionaries
do not change whimsically, but they do evolve.
In science, however, the more explicit the definition for a concept,
the better we can determine when it applies and when it does not
apply. A scientific concept can mean only what a community of interacting
scientists agrees that it means. Although scientists sometimes
argue about the “proper” meaning of concepts, a theory tends to gain
a footing in the scientific community once scholars settle on meanings
for the time being. Many concepts in scientific theories refer to states
of affairs with fairly stable properties. For example, we cannot have a
coherent theory about the distribution of housework responsibilities
if the meaning of housework changes every time we use this word.
Many but not all concepts can correspond to a set of empirical
measures. A variable is any measure that can have two or more values
such as yes or no, strongly agree to strongly disagree, or even a range
of values such as the Centigrade scale of temperature. For example, we
may measure housework by hours of work within the house or by a
specified set of tasks such as doing laundry, toilet cleaning, and so on.
Variables that might measure housework could be the hours a person
spent doing the work, the economic value of the work, and so on.
Relations play the role of verbs in a theory. To relate concepts to
each other, we need connecting relations asserting how the concepts
are related. All relations have properties (symmetry, reflexivity, and
transitivity). So for example, we could use the same concepts but
with different relations, and the truth of the statement will change
In these two cases, one relation is transitive (is a relative of), and
the other is not transitive (is the father of). This simple difference
in the properties of relations makes all the difference in the validity of
the final statement. Relations can often be formalized in simple terms
such as greater than or less than, and definitions are signified by identity
or equals. As we shall see shortly, this fact provides a key link
between theory and research.
Propositions exist when a concept is linked-in -a meaningful way
by a relation to another concept. So we could say that among dualearner
couples, the social class status of the husband is positively
related to the amount of housework he performs. This would be a
theoretical proposition. The first concept is social class status of the
husband. The second concept is the amount of housework. The relation
in this proposition says that the greater a husband’s social class status
(relative to other husbands), the greater the husband’s amount of
housework. This relation can be mathematically expressed (modeled)
as a function:
Amount of housework = f (Social class status).
As we shall see shortly, our first approximation of this functional
relation (f) between a husband’s social class status and amount of
housework is usually a straight line with either a positive or negative
slope. When a proposition asserts covariation between variables, the
relation between concepts includes a sign (positive or negative), indicating
that increases either occur together or go in opposite directions.
When a relation asserts causal influence, the relation also makes clear
which variable (independent) is influencing which other variable
Earlier in this chapter, we defined a scientific theory as “a set of
systematically related propositions that are empirically testable”
(Rudner, 1966). Now that we have the building blocks in place we can
discuss the meaning and implications of this definition in greater detail.
Systematically related propositions. A theory usually comprises several
propositions. Indeed, one proposition alone would not constitute a
theory but would simply be a conceptual “hypothesis.” Not only must
a theory have at least two propositions, but these two propositions
must be systematically linked by relations. The way we link one theoretical
proposition to another is by logical form. Although there are
many mathematical and logical systems at our disposal, we will use
simple syllogisms to show what we mean by form and relations. Note
that the mini theory below is used only for an example of logical form
and should not be regarded as empirically adequate.
Imagine that we have two propositions, but they may at first
appear unrelated. For example, from a set of propositions about the
intergenerational transmission of social class, we find the following:
The greater the family of orientation’s value on education, the greater the
son’s social class status.
If we combine this with our previously discussed proposition, we get
a mini theory with two related propositions. Furthermore, we can
deduce a third, new and interesting proposition from these two.
The greater the family of orientation’s value on education, the greater the
son’s social class status.
The greater the husband’s social class status, the greater the amount of
housework performed. Therefore, the greater the family of orientation’s value on education, the
greater the amount of housework performed by the son (husband) in his
family of procreation.
Now the above mini theory is in the form of a syllogism. These
deductive arguments are constructed so that if the first two propositions
(called premises) are true, then the deduction, called the
conclusion, is necessarily logically true (although not necessarily
empirically true). Much more important is that if the conclusion is
false, then we know that at least one or both of the premises are also
false. These properties hold only if the correct form is followed.
In regard to syllogisms, the correct form is that the middle term B
(below) must appear at the end of the first premise and at the
beginning of the second premise. This is called the rule of distributed
Regardless of whether these ideas are in the correct form, what we
want to know is if our theory as a whole is accurate. In logic, if we
showed that our conclusion is true, that would not mean that the two
premises are also true. Indeed, when we make such an assertion, it is
called the “fallacy of ‘affirming the consequent.” In reality, we have to
work backward. What we try to do is to disprove our conclusion. If
we show that our conclusion is false, then we also know that at least
one of the premises is false. We get much greater intellectual power
out of using this reverse logic (modus tollens). This is one of the major
theoretical reasons we use the null hypothesis in testing our
Empirically testable. For a set of propositions to qualify as a scientific
theory rather than religious or literary or political, it must be capable
of being empirically tested. That does not mean that all our scientific
theories have been empirically tested but that a theory that is incapable
of being empirically tested is not a scientific theory. Now we
turn to the question “What does it mean to say that a theory is
capable of being empirically tested?”
The proposition in the conclusion of our mini theory used above
can be used as an example. It states that the degree to which the
husband’s family of orientation values education is related to the
amount of housework the husband will perform in dual-career couples.
This concluding proposition will now become our research focus. When
we decide to research a particular proposition, it becomes the “conceptual
hypothesis” in addition to being the deduced conclusion of a
The first thing we must do to research this “conceptual hypothesis”
is to find ways to measure its components as variables. That
means that we must find or develop measures for the first concept,
husband’s family of orientation’s value on education, and the second
concept, amount of housework performed. Finally, we must not forget
that we must find a way to demonstrate (or operationalize) the
relation between these concepts. We might operationalize or measure
the family of orientation’s value on education by asking the husband
several questions about his perception of his mother and father’s
values about education and how much financial assistance they offered
for his education. Both of these measures would have problems of
validity and reliability because of faulty recall, halo effects, and so on.
We might measure amount of housework by either the number of
hours spent on various household tasks or the economic value of tasks
performed by the husband. Both of these measures would have problems.
Which tasks should we list? Would the husband inflate his estimate
of the time spent? If we had the husband list the tasks and then
assigned the economic value at the cost of a plumber for plumbing
jobs, a babysitter for child care, and so on, would we find that this
correlates poorly with the measure of amount of time spent on household
chores? Indeed, there are always problems and questions regarding
how accurately our empirical operationalizations measure our
concepts in our conceptual hypothesis.
The last component of our conceptual hypothesis to be operationalized
is the relation. As we indicated previously, we hypothesize
that amount of housework is a function of family’s values on education.
What is this function? In most social research, the first functional
form that we test is that of a straight line. You may recall the slopeintercept
formula for a straight line (y = b [x] + a). Indeed, this simple
mathematical model is the basis for much of our analysis of relations
in the social sciences. The formulae for a straight line is a mathematical
model and it has a corresponding statistical model expressed in
the formula (Y = a + b [X]+e). Basically, the application of the statistical
formula in the analysis of the data allows us to examine the linear
relation between amount of housework (y) and the family’s value on
education (x). The strength and direction (positive or negative) of that
relation will be given by the regression (correlation) coefficient (b).
Concept 1 Relation Concept 2
Family of orientation’s value on education Husband’s amount of housework
Husband’s perception of family’s value Husband’s reported hours of housework
Measure 1 Relation Measure 2
Figure 1.1 Conceptual and Measurement Hypotheses
Once all of the measures are in place, we have in effect two
hypotheses. Our theoretical proposition represents the conceptual
hypothesis, and our measures and the relation between them represents
the measurement hypothesis. The architecture of the empirical
test of a theory is portrayed in Figure 1.1.
Finally, if we were to conduct our study on husbands in dualcareer
couples, we would have to control for several variables that
might account for the relationship, such as number of children, maid
service, wife’s level of housework, and income level. After we analyze
the results, we might find that there is no relationship between the
variables we measured. If we can satisfy our critical colleagues that we
have valid and reliable measures of the concepts and relations in the
conceptual hypothesis, then we could conclude that the theoretical
proposition is false. Because this is a deductive system (syllogism), we
would also know that at least one of our premises is also wrong. As a
result, it would be time either to recast and modify the theory significantly
or even to discard it completely.
On the other hand, what if our results are supportive of the
conceptual hypothesis? If we can satisfy our colleagues regarding the
validity and reliability of our measures and procedures, then we might
want to say that the theory is true. However, as we said previously,
this would be committing the fallacy of affirming the consequent. If
the proposition seems to hold, it does not mean that the other propositions
(premises) are true because it is logically possible to deduce
the conclusion from several other sets of premises. We can never
determine which possible set of premises might be true. As a result,
science and theory achieve a great deal by disproving hypotheses and
theoretical propositions rather than ever proving anything. All our
scientific knowledge is tentatively held until we disprove it. Scientist
are not true believers, but skeptics.
FUNCTIONS OF THEORY
Scientific theories serve many purposes. Here, we emphasize how
they contribute to understanding. It is widely recognized, however,
that knowledge for its own sake is not the only worthwhile goal in life.
We all want to do something with our knowledge. In our preceding
example, if we had some confidence in our theory about housework, a
young woman searching for a helpful mate might want to examine the
educational values of the family of orientation of prospective spouses.
More seriously, for those in one of the helping professions, theories
about the family may provide knowledge to improve their services as
therapists, program evaluators, or social policy advisers. For those
who are political activists, agitators for modest or revolutionary causes
or the reduction of major social problems, or even staunch defenders
of the world as it is or once was, theoretical knowledge about families
can be put to fruitful use. Without meaning to diminish the importance
of any of these practical uses of theories, we concentrate here on the
ways in which theories contribute to the immediate goals of science.
1. Accumulation. Theories assist in the accumulation and organization
of research findings. Much of the pursuit of knowledge involves
the collection and analysis of empirical facts, filtered through the
lenses of researchers. A body of empirical knowledge, however, is just
a pile of findings. Theories tell us how to select and arrange research
findings into meaningful groupings. If no existing theory is available
for this purpose, new theories can be constructed from findings
through a process of tentative generalization.
2. Precision. Theories articulate ideas in more carefully specified ways
than everyday language allows. Thinking theoretically forces one to
clarify what concepts and relations really mean and what they include
and exclude. This precision facilitates communication, so long as the
communicators are trained in the language of science.
3. Guidance. Theories direct researchers to develop and test measurement
hypotheses (i.e., empirical statements about what the data are
expected to look like). Because theoretical ideas entail abstract,
plausible, and tentative arguments, they must be checked against the
empirical evidence for confidence in them to grow. Theories point to
new kinds of relevant evidence for which findings do not currently
exist. It is relatively easy to find or create evidence in support of a
theory that one likes, however. Theories, therefore, also promote a
critical spirit of inquiry. Efforts are made to refute theories, not just
to support them. If a field contains two or more theories that yield
incompatible hypotheses, it may be possible to design a study or a
series of studies that help to decide which theory is better.
4. Connectedness. Theories demonstrate how ideas are connected to
each other and to other theories. Theories are systematic sets of ideas.
The parts of a theory fit together in a coherent way. Knowing what
the parts are and how they fit together helps to distinguish one theory
from other theories. This also helps us to see what two or more
theories share or how they could be connected or combined by bringing
together elements of each.
5. Interpretation. Theories help to make sense of how the phenomena
they cover operate. There are at least two aspects of sensibleness in
interpretation. One is essentially descriptive: A theory should enable a
good description of the subject matter with which it is concerned. This
does not necessarily mean that the theory must fit our intuitive judgments
of the way things work. In fact, a theory may challenge many
intuitions or commonsense views of the world. Rather, to enable us to
describe a subject, a theory provides a plausible picture of structures
or processes that we can accept as reasonable given the assumptions
in the theory. The other feature of interpretation is that theories evoke
or promote stories about the way things work. If we can visualize
concrete or even general scenarios of how things work, the theory is
providing an interpretation.
6. Prediction. Theories point to what will or can happen in the
future. If a theory helps us understand what has happened in
the past or is happening now, this is of course desirable. To continue
to be useful, however, a theory must contain propositions equally
applicable to future events and experiences. This does not mean that
any theory accurately foretells the future, only that it should make
some relevant propositions regarding future outcomes. If we know
what a theory predicts, this can contribute to the guidance function
mentioned earlier. If a theory predicts something that is subject
to human intervention and alteration, future actions may invalidate
the theory. Normally, however, a theory is expected to hold up
in the future to the extent that its predictions are confirmed.
Even if predictions are not confirmed, the theory may be basically
correct, and only the conditions under which it works may require
7. Explanation. Theories provide answers to “why” and “how” questions.
As the quotes in Box 1.1 from Burr (1973) and Homans (1964)
suggest, explanation is often considered the single most important
function of a theory. Because explanation is so central to the functioning
of theories and because it has a variety of different meanings, we
need to address this topic in more detail. Without theories, we cannot
determine why and how things happen the way they do. In our everyday
lives, we may be satisfied by saying that one specific event, say, a
divorce between two friends, happened because of one or more other
events that took place before the divorce. Such reasoning captures
some of what goes into a scientific explanation, but it overlooks the
most important aspect. Scientific theories explain by relying on deductive
arguments. Specific events and the connections between them must
be derived from more general statements for us to say that the events
and the connections between them have been explained.
Explanation is provided by a general statement (sometimes called
a “covering law”) that includes (or covers) the specific instance we
want to explain. So for example, if you drop this book and it falls to
the floor, we could ask you to explain why it fell. You would simply
cite the law of gravity as explaining why the book fell. This is to say
that the particular phenomenon is part of a generally understood pattern
we call “gravity.” Now this is not to say that in reality our understanding
of gravity is simple; indeed, it involves complex theories of
electromagnetism, mass, space, and time. But the “law” derived from
these complex theories covers the specific phenomenon of the falling
book. Thus, to explain why someone got divorced (a particular
instance) would be to provide the general theoretical propositions
(covering laws) explaining divorce.
It is crucial to remember that although functions or goals like
those we have listed represent ideals, this is no guarantee that a particular
theory will fulfill such ideals. In fact, we can use the seven listed
functions as standards against which to measure the performance of a
theory at any stage of its development. A theory may perform well in
some respects but less well in others. Determining such facts helps us
decide where our energies need to be directed to improve the theories
that we have or invent better ones.
Ideas about human families can come from many sources, including
scientific disciplines concerned with the study of entities that are not
families, as well as personal experiences with our own families. No
potential source should be dismissed, although we should be aware
that those sources selected likely will affect key features of the theory
that emerges. Most of the family theories that exist today draw on
ideas from external sources. Family theories are not insulated sets
of ideas, and family theorists do not merely talk to each other about
For a theory to be about families, there must be at least one
family concept in the theory. We cannot decide what a family concept
is, however, unless we first decide what a family is. Let us, therefore,
begin by thinking of a family as a social group. We need to identify
the distinguishing features of this group. Following are some of the
major ways that families differ from such groups as associations of
coworkers and networks of close friends.
1. Families last for a considerably longer period of time than do most
other social groups.
Of course, some relationships in families are not enduring.
Marriages can be broken by divorce or death fairly soon after they
are formed. Yet we normally think of our own families as lasting
throughout our lifetimes. We actually are born into a family that already
exists. Our parents remain parents even after we become adults. We add
members to the family when we marry and become parents. Our siblings
remain siblings throughout our lifetimes.. Although it is possible
for coworkers and close friends to maintain relationships for long
periods of time, families are the only groups that virtually require lifetime
membership, even though some members are added and subtracted
along the way. Belonging to a family is involuntary in the sense
that we do not choose which parents are going to give birth to us. Other
groups tend to be much more voluntary, in that we have some choice
about joining them in the first place.
2. Families are intergenerational.
Through the act of giving birth, families include people who are
related as parents and children. If elders live long enough, we have
ties to grandparents, and maybe even to great-grandparents and
great-great-grandparents. At some point, we ordinarily have living
members of both older and younger generations, and we may eventually
become grandparents or great-grandparents ourselves. Other
kinds of groups may include people with fairly large age differences,
but families are the only groups that virtually guarantee this.
The fact that the human infant at birth is a helpless creature and
cannot approach self-sufficiency for almost 20 years means that the
intergenerational bond is particularly crucial to human survival. Every
child needs some sort of caretaker and caregiver, whether it is a
biological parent, an adoptive or foster parent, or somebody else who
takes the responsibility for providing nurturance during the early
years of life. It is no accident that our image of family includes an
3. Families contain both biological and affinal (e.g., legal, common law)
relationships between members.
It is the biological act of birth that creates the fundamental
family tie. This act also means that we share at least some inherited
characteristics or proclivities with family members that are directly or
indirectly related to us by birth. At least until humans perfect the
cloning of adults, and perhaps even then, the process of becoming a
person will be based to some degree on biology. Families are in the
business of producing and sustaining persons and personhood. Even
though work groups or friendship groups may sometimes contain biologically
related members, such groups tend to have other purposes.
There is also a social side to this process of creating persons. No
society leaves the biological act of birth or the rearing of children to
chance. Personhood is achieved through a process of socialization.
That socialization is subject to secular and religious rules about how
the process should be carried out. Pursuant to these rules, family
members have rights and obligations, which tend to be codified in
both laws and informal agreements.
Aside from adoption, the major legal provision about families
concerns marriage. We may not think of a marriage in itself as
constituting a family, but we recognize marital relationships as part of
families. Some families are conjugal, in that they contain one or more
marriages. It might even be argued that if humans didn’t have families
they wouldn’t need marriages, although families may often function
well without marriage. In any case, marriage itself involves rights and
obligations under the law, and it also creates family ties in law. Some
of our family members join and leave the group either because of our
own marriages and divorces or because of the marriages and divorces
of other family members.
Other kinds of groups are subject to regulation by laws (e.g.,
contracts) and informal agreements, of course. Such regulations may
exclude as well as include people in work and friendship groups, and
they govern proper conduct within such groups. What such groups do
not have, however, are relationships anything quite like, for example,
cousins or aunts and nephews, which arise because our mother’s sister
4. The biological (and affinal) aspect of families links them to a larger
It follows from what we already have said that families are not
just small groups of closely related individuals who live together or
interact on a frequent basis. Families extend outward to include
anybody sufficiently related to us by blood, marriage, or adoption.
This kinship group may have the identifiable boundaries of a clan, or
it may be loosely organized and diffuse. Everybody stops counting
distant relatives as family members at some point along the periphery.
Nevertheless, the ties of kinship create the potential for lineages and
collateral (i.e., within generation) family relationships that can
become quite extensive. Through kinship, families are tied to history,
tradition, and multiple generations of group members. In some societies,
these kinship groups are major features of the social, cultural,
political, and economic landscape. Work and friendship groups tend
to be much more temporally and spatially encapsulated.
To answer the questions in Box 1.3, we likely will find ourselves
asking other questions. Do we visualize the people listed in each
example as all being part of one family, or do they belong to two or
more different families? Are the listed persons the only ones in the
family or could there also be others (i.e., is this a whole family or just
part of one)? Should we think of these persons as members of an
actual family, or are they a family in either a symbolic sense or in the
way they function in their relationships? Are we thinking of how
common or uncommon a grouping is in a particular society, either in
frequency or in what is expected as normal, natural, and acceptable
by law or custom? And are both affinal and consanguineal relationships
identified as family relationships?
We could discuss additional arguments about the distinctive
features of families as groups. For example, we could draw attention
to certain qualities of social interaction that are commonly found in
families but not elsewhere. One of these is that family members
supposedly care about one another as whole persons (cf. Beutler,
Burr, Bahr, & Herrin, 1989). Whichever criteria are considered,
however, the distinctiveness of family groups tends to be only a
matter of degree. Nonfamily groups, such as networks of friends or
coworkers, usually have some family properties but fewer of these
properties or in less obvious amounts.
Box 1.3 Which of These Is a Family?
A husband and wife and their offspring.
A single woman and her three young children.
A 52-year-old woman and her adoptive mother.
A man, his daughter, and the daughter’s son.
An 84-year-old widow and her dog, Fido.
A man and all of his ancestors back to Adam and Eve.
The 1979 World Champion Pittsburgh Pirates (theme song: “We Are
Three adult sisters living together.
Two lesbians in an intimate relationship and their children from a previous
marriage of one woman and a previous relationship of the other
woman with a male friend.
Two children, their divorced parents, the current spouses of their
divorced parents, and the children from previous marriages of their
A child, his stepfather, and the stepfather’s wife subsequent to his
divorce from the child’s mother.
Two adult male cousins living together.
A 77-year-old man and his lifelong best friend.
A childless husband and wife who live 1,000 miles apart.
A widow and her former husband’s grandfather’s sister’s granddaughter.
A divorced man, his girlfriend, and her child.
Both sets of parents of a deceased married couple.
A married couple, one son and his wife, and the latter couple’s children,
all living together.
Six adults and their 12 young children, all living together in a communal
It also is worth noting that we can conceptualize the family not
only as a concrete group or social organization but also as a social
institution. As an institution, family includes all of the beliefs and
practices of and about all of the families in a particular society or
geopolitical context. It also includes the ways in which different
families are connected to each other and to other social institutions.
For example, in most modern and well-developed societies, families
tend to have a private character, each one walled or fenced off from
other families and from public view much of the time. There also are
expected and experienced linkages between families and schools,
families and the workplace, families and governments, families and
the mass media, and so on. Because members of families often share
economic resources and collaborate in both productive labor and the
consumption of resources, it is also useful to consider families as being
involved in a society’s system of social stratification. Thus, some
families have more wealth, power, and status than do other families.
With an idea of the meaning of family, we can begin to envision
what counts as a family concept. Some of these concepts describe the
composition, or size and configuration, of family membership. Some
describe the structures and processes of interaction that take place
between family members. Some describe the ways that families relate
to their environments. Some will describe the whole family as a group,
the family as an institution, or the nature of the ties between two or
more members within the family group.
We believe that for a theory about human relationships to be a
family theory, one or more ideas (concepts or variables) about
families must be included in the theory. The family idea may appear
in either or both of two places: (a) as part of the explanation (e.g.,
parental discipline helps explain juvenile delinquency), (b) as the phenomenon
to be explained (e.g., the state of the economy in a country
helps explain the divorce rate in that country), or (c) both (e.g.,
marital communication styles help explain marital satisfaction).
Thus, we can distinguish two general kinds of family theories, those
that are “about the family” or explaining how families work (b and c
above) and those that consider family ideas to be useful explanations (a
and c above). Of course, a wide variety of ideas, some familial and some
not, may be needed in combination to adequately explain family life
(e.g., marital affection plus the economy influence the odds of divorcing).
Likewise, the forces that help explain family life may also help to
explain other things (e.g., the state of the economy may influence stock
prices as well as the divorce rate). When assessing the family theories
of others, or when creating one yourself, it helps to locate where the
family ideas are or at least where you think they belong.
A Brief History of Theory in Family Studies
Each theory in family science has its own historical legacy, as we
discuss further in subsequent chapters. Our focus here is on major
themes and examples of how family theory has been important to
scholarship throughout history. We refer to several useful accounts of
this history and recommend them for greater depth of coverage.
Adams and Steinmetz (1993) surveyed contributions of the
classics, which encompass ideas up to the 20th century. Early philosophers
were often interested in prescribing ways of living according
to their own values. Adams and Steinmetz called attention to this infusion
of ideology, and they noted that it sometimes was accompanied
by efforts to describe families in more dispassionate ways. Only
scattered references to family life per se are contained in the classical
works, with little resemblance to the kind of scientific explanation we
have now come to expect of scholarship.
Much of the early work centered on attempts to find the origins
of marriage as an institution and to trace its evolution as societies
moved toward a more modern form (Adams & Steinmetz, 1993,
pp. 76-78). Some scholars searched for the ultimate purposes of
marriage or the family. Some saw progress or at least the adaptation
of family life to changing social circumstances. Some lamented the
declining importance of the family, especially as industrialization and
urbanization gripped the Western world in the 19th century. Adams
and Steinmetz (1993, p. 86) conclude that the closest these social
philosophers came to a real theory was a model of parental socialization
that resulted in positive outcomes for children and the meeting
of needs for both generations when parents became elderly. For the
most part, however, and whether obvious or subtle, ideology in the
classics, Adams and Steinmetz (1993) found, is seldom totally absent,
although its presence makes a theory neither right nor wrong (p. 93).
Howard (1981) examined the history of U.S. family sociology
from 1865 to 1940. He also noted the early emphasis on evolutionary
thinking, as well as the interest of “moral reformers” in doing something
about the problems families seemed to be facing because of the
industrial revolution. Howard considered the period between 1890 and
1920 a progressive era. Even as moral reformers and charity workers
expanded their initiatives, the idea that families could adapt to changing
environments became popular. An emphasis on the psychosocial
interior of family life took hold, and educational programs to foster
the socialization of children seemed promising. If families were sometimes
struggling and disorganized, it was because they were caught
between two conflicting value systems, one emphasizing traditional
images of the family and the other based on the requirements of the
modern democratic and capitalistic state.
The years 1920 to 1940 constitute the latest period Howard
described in detail. In these years, the key theme in family scholarship
shifted from ecology to interaction among family members, with the
goal of personal adjustment. In 1924, the American Sociological
Society (now the American Sociological Association) established its
Family Section, and at about the same time, Ernest Groves developed
the first systematic college course in family life education at Boston
University. Howard also noted as a counter-theme during this period
a renewed emphasis on the institutional level of analysis, with studies
of families in a community context and attention to cultural diversity
on a broader scale. In a chapter published with Howard’s book, van
Leeuwen (1981) noted that European family scholars remain more
interested in the institutional, or macroscopic and historical, levels.
On both continents, however, the norm developed that family scholars
should refrain from moral and political evaluations as they increasingly
relied on empirical data collected through fieldwork (van
Leeuwen, 1981, p. 133).
Reflecting on much of the same body of work mentioned above,
Christensen (1964) concluded that systematic theory building in the
family field did not begin until about 1950. Concepts and rough orientations
were taking shape in the first half of the century, along with a
growing industry of empirical research, but without formalized explanations
meeting the requirements of propositional theory. Although not
highlighted in any of these historical accounts, an upswing in scholarly
interest about families within many academic disciplines appeared
before 1950, with contributions by sociologists, anthropologists,
psychologists, home economists, and social workers, among others. Yet
Christensen’s designation of 1950 as a turning point seems to be
basically correct. An inspection of the International Bibliography of
Research in Marriage and the Family, 1900-1964 (Aldous & Hill, 1967)
shows that of almost 4;000 entries before 1950, only 7 contain theory or
a cognate term in their titles. Of the 12,000 entries for the 1950-1964
period, 93 entries contain such terms. By comparison, for the 2-year
period from 1991 to 1993, 264 of 7,600 entries in a subsequent inventory
(Touliatos, 1994) pertain to family theory. Interestingly, the largest
proportion of theoretical works before and during the 1950s dealt with
courtship and mate selection. This was the first topic to receive systematic
and cumulative theoretical treatment.
Aside from the increasing attention to family theory in the scholarship
of the field since 1950, the last half of the century can be
subdivided into three stages, each with its own set of themes.
CONCEPTUAL FRAMEWORKS: 1950-1966
The single most prominent theme in family theorizing during the
1950s and early 1960s was an emphasis on identifying conceptual
frameworks. This emphasis is evidenced by a series of works devoted
to the topic (Christensen, 1964; Hill, 1951; Hill & Hansen, 1960;
Nye & Berardo, 1966/1981). The number and character of the
particular frameworks varied, but the basic idea remained the same.
As research on family life accumulated, most analysts attempted to
give explanations for their findings. The explanations often were
narrowly focused, and it was difficult to see how various studies fit
together. The attempt to identify frameworks was a search for underlying
principles that might help in the construction of general theories
for the field. It was not claimed that such theories already existed, nor
was it often assumed that a single integrated theory would be practical.
Rather, the idea was that by comparing the currently fragmented
works with respect to the concepts and assumptions they used, scholars
might be able to work toward the building of family theories in a
more coordinated way.
Christensen (1964) captured the spirit of this period well in the
closing remarks of his introductory essay in the Handbook of
Marriage and the Family:
As has been said several times, there is urgent need for better theory.
Critics of family research have described it as being amateurish,
trivial, scattered, often sterile, and sometimes moralistic.. . . There
is still need for correcting what Goode (1959, p. 186) spoke of as a
hornet’s nest of conceptual and terminological problems. There is
still need to isolate and then integrate, insofar as seems feasible, the
theoretical frameworks which can guide the discipline. And there is
still need to find and then specify the relationships among empirical
generalizations in order to constitute true theory. (pp. 29-30)
FORMAL THEORY CONSTRUCTION: 1967-1979
A change in tone characterized the next several years of work on
family theory. Following Hill’s (1966) address after receiving the first
Burgess Award for a career of scholarly achievement in the family
field, attention turned toward methods of deductively and inductively
creating theories, using a clearly delineated propositional format.
Many examples of this type of work emerged. Nye and his colleagues
presented a propositional theory of family stability (Nye, White, &
Frideres, 1969), followed by a propositional theory of age at marriage
(Bartz & Nye, 1970). Goode and his colleagues published a massive
volume listing hundreds of propositions relevant to scores of family
topics (Goode, Hopkins, & McClure, 1971).
The most important works of this period, however, were those
spearheaded by Burr (Burr, 1973; Burr, Hill, Nye, & Reiss, 1979,
Vols. 1 & 2). In his 1973 volume, Burr applied the principles of
deductive, propositional theory building provided by philosophers
and sociologists to 11 topical areas of research. In the first volume of
Burr, Hill, et al. (1979), Burr and his colleagues applied the same
procedure to twice as many areas, involving experts in those areas as
chapter authors. In Volume 2, Burr, Hill, et al. (1979) focused on five
general theories that were not substantively limited. The editors
noted that the second volume was not comprehensive in its coverage
of general theories, and they acknowledged difficulties in linking the
two volumes. Inductively integrating materials from the first volume
proved difficult because of the lack of semantic equivalence across
domains, the complexity of models in the first volume, and the lack of
a way to bridge macro- and microlevel propositions (Burr, Hill, et al.,
1979, Vol. 2, pp. xii—xiv).
Organizational developments within the National Council of
Family Relations (NCFR) also were important during this period. In
his Burgess address, Hill (1966) had called for the creation of a
Theory Section to parallel the existing Research Section. Although his
initiative was defeated, by the late 1960s NCFR had created the joint
Research and Theory Section that continues today. Hill did not relent
in his effort to give visibility to theorizing as a major professional
activity, however. With the help of Nye and others, he created the
Workshop on Theory Construction, which began with meetings at the
NCFR annual conference site just before the regular NCFR conference.
The workshop served as a training ground for both students and
more advanced family scholars, allowing them to nurture and demonstrate
their talents in theory building. In 1975, the name of the workshop
was expanded to ‘Theory Construction and Research Methodology.
This workshop also survives today. In its first 24 years, more than
560 papers were presented, and more than 730 participants appeared
on its programs It is interesting that both the section and the workshop
evolved to combine theoretical and research interests. This
symbolizes the extent to which family scholars see the two enterprises
as fundamentally connected.
PLURALISM: 1980 TO THE PRESENT
It is always difficult to gain perspective on developments in a
field when they are still swirling and the next major turning point is
unforeseen. Lacking the distance afforded by time to reflect backward,
we tend to see the present as chaotic or transitional. Of course, this
perception also may mean that our tendency to periodize the past is
an oversimplification, and all moments of intellectual history (and
even family history) are fuzzier and more complex than they appear
in retrospect. Nevertheless, the publication of the two edited volumes
by Burr and colleagues (Burr, Hill, et al., 1979) seems to represent the
high-water mark, if not the culmination, of the theory construction
The pluralism of the most recent decades has several possible
interpretations. It can be seen as a continuation of the quantitative
growth of contributions by scholarly participants in family theory and
research. It can also be taken to be a furthering of fragmented and
specialized interests. Or it can be viewed as a new respect and tolerance
for diverse philosophies, theories and theory-building methods,
and research strategies.
Doherty et al. (1993) provide a cogent summary of nine main
trends, although they gave these developments somewhat more recent
origins than we would. We excerpt their list from a more detailed
1. The impact of feminist and ethnic minority theories and perspectives
2. The realization that family forms have changed dramatically
3. The trend toward greater professional (multidisciplinary) inclusiveness
4. The trend toward more theoretical and methodological diversity
5. The trend toward more concern with language and meaning
6. The movement toward more constructivist and contextual approaches
7. An increased concern with ethics, values, and religion
8. A breakdown of the dichotomy between the private and public
spheres of family life and between family social science and family
9. Greater recognition by family scholars of the contextual limits of
family theory and research knowledge (pp. 15-17)
To be sure, Doherty et al. (1993) see challenges posed by, and
potential problems with, these emergent developments, but their basic
posture is laudatory and optimistic. Nonetheless, any idealistic hope
on the part of some leaders in the field that we would eventually
converge around one grand theoretical scheme to explain everything
about families has now been completely dashed and discarded. What
we see is the continuing proliferation of theories, the eclectic combination
of elements from different theories, and variations and transformations
within existing theories. It is as common now, if not more
common than ever before, for theorizing to be narrowly focused on
particular topics and issues. Countless mini theories, middle-range
theories, and models of mostly causal processes, all closely linked to
Hempel’s (1952) plane of observation, characterize the literature,
particularly in professional journals and in the proceedings of conferences
The other main trend in recent years has been a vigorous questioning
of the philosophical foundations in the field (cf. Osmond,
1987; Thomas & Wilcox, 1987). The result has been a turn toward
interpretive and critical philosophies (see Box 1.2) and away from at
least some elements of the positivistic philosophy. It is important to
remember, however, that the entire scholarly community does not
move in unison in any one direction. Some theorists attempt to blend
and integrate, whereas others find one philosophy or even one theory
compelling and the others inadequate. The current scene is a curious
and shifting mixture of consensus and conflict over theories and
In light of the current situation, a precaution is in order as we
prepare to explore in some detail seven of the currently important
theoretical frameworks in the family sciences. Our own map of the
alternatives is not the only possible one, nor is it necessarily complete
and otherwise adequate. In 1950, only the symbolic interactional
perspective, among the theories we cover, would likely have been
widely recognized. Until the mid-1960s, three other perspectives that
we do not cover as distinctive were commonly identified (structurefunctional,
institutional, and situational). By this time, the developmental
perspective had gained recognition, but several others were
either widely rejected or absorbed into other points of view. When
Nye and Berardo (1966/1981) attempted to comprehensively cover
conceptual frameworks at the end of this period, they even added six
more, each roughly representing an academic discipline (i.e., anthropology,
psychoanalysis, social psychology, economics, legal studies,
and Western Christian studies). When Broderick (1971) reviewed the
theoretical developments of the 1960s, he saw several new possibilities.
Of these, the exchange and systems perspectives have continued
to thrive, and they are represented in our following chapters. By the
end of the 1970s, a similar review by Holman and Burr (1980) added
conflict, behavioral, ecological, and phenomenological perspectives
as being of at least secondary importance. Of these, we give major
attention only to the conflict and ecological perspectives. When Nye
and Berardo (1966/1981) reissued their 1966 book, they added
exchange, systems, and conflict theories in their introduction, as well
as “social individualism” and “transactional analysis,” neither of
which seem to have progressed thereafter. Finally, the most recent
comprehensive treatment includes more than a dozen chapters on
distinct theoretical perspectives (Boss et al., 1993), with the notable
addition of feminist theory.
In a study (Klein, 1994) of more than 100 family scholars, the
participants were asked to rate 11 alternative theories as to how
favorably disposed they were toward them. Of those we emphasize in
this book, their ranking, in descending order was as follows: symbolic
interactional, developmental and systems (tied for second place),
exchange, conflict, and ecological. In general, however, the participants
indicated dislike for very few, and most were favorably disposed
toward several. Furthermore, the scholars nominated dozens of others
that they liked, although the nominations seldom coincided. We conclude
from this pattern of findings that no single scheme focused on
only a small number of general theoretical perspectives would be
A fruitful typology of theories is only partially in the eye of the
beholder. Some threads seem to be common to the various alternative
classifications, and some reflect historical changes in the field regarding
the popularity and promise of the alternatives. Nevertheless, each
mapping of the territory is at least somewhat dependent on the purposes
and viewpoint of the author. In our case, we aim for relative
simplicity to meet the needs of readers. We capitalize on the existing
complexity mainly by covering some of the varying and more recent
emphases within the broader traditions. We use an organizational
concept throughout this book: that of “theoretical framework.” What
we mean by theoretical framework is that there exists in each scientific
field a core of theoretical assumptions and propositions. We do
not simply refer to these as theories, because although they are theories,
they also are sufficiently general to have given rise to a number
of theoretical variants. In this one sense, these theoretical frameworks
are like families, with the core theory providing sufficient richness for
the development of offspring. After exploring these various frameworks,
we return in chapter 9 to the problem of fitting the theories
into a coherent overall picture.
Adams, B., & Steinmetz, S. (1993). Family theory and methods in the classics. In
P. Boss, W. Doherty, R. LaRossa, W. Schumm, & S. Steinmetz (Eds.), Sourcebook
of family theories and methods: A contextual approach (pp. 71-94). New York:
This is an excellent historical overview of the development and evolution of
What Is a Theory? 31
Beutler, I., Burr, W., Bahr, K., & Herrin, D. (1989). The family realm: Understanding
its uniqueness. Journal of Marriage and the Family, 51, 805-816.
This is a bold attempt to identify the distinguishing characteristics of families. The commentaries
by scholars that follow on pp. 816-829 indicate the controversy surrounding
the proposals of Beutler and his colleagues.
Burr, W. (1973). Theory construction and the sociology of the family. New York: John
This book, despite its age, remains an excellent introduction to theory construction and
philosophy of science. The first chapter is outstanding.
Thomas, D. L., & Wilcox, J. E. (1987). The rise of family theory: A historical and critical
analysis. In M. B. Sussman & S. K. Steinmetz (Eds.), Handbook of marriage
and the family (pp. 81-102). New York: Plenum.
This provides a solid historical picture from the perspective that the field is moving to
a less positivistic approach.
pete padillla sociology