Family Theory

Suppose a couple is recently divorced. A friend may wonder why this happened and develop several hunches. Perhaps they argued a lot, and one or both may have frequently seemed upset. The friend may begin thinking about why the couple got married in the first place. Perhaps their dating relationship was unusual, or perhaps their upbringing as children offers clues. Using information about their past, the friend might develop a theory, a speculative argument about factors contributing to the couple's divorce. The word theory derives from the Greek verb theorein, meaning to behold or contemplate. People have contemplated the nature and operation of human families at least since the ancient Greeks, and they continue to do so today. All individuals may wonder how their own or other families work and about the problems involved.

Theorizing about a particular event or a particular marriage or family seems natural in everyday life. Social scientists, however, are not interested in explaining single events or how one marriage or family works. Instead, social scientists want to know how marriages and families work in general. This does not mean that every divorce will have the same cause, only that the emphasis is on a broad understanding of many marriages and families. If they know what generally is true by examining many different marriages and divorces, they come closer to developing a useful theory about divorce. A useful theory provides a general understanding of what has happened in the past, and it enables the scientists to make predictions about what might happen to other couples in the future.

Furthermore, if social scientists want to help other couples deal effectively with their relationships, they need to have confidence that information about a particular couple is not unique. They need to know what makes marriages similar to one another and especially what makes some marriages different from others.

Why is it impossible to have a scientifically useful theory about one event, such as a particular divorce? Suppose it is strongly believed that certain factors in a couple's past are responsible for their divorce. To be sure of this, the social scientists would have to argue that if the couple's pasts had been different in certain ways then they would not have gotten divorced. Something that actually did not happen might have prevented the divorce. The problem is that scientists cannot know about things that did not happen. Such unknown circumstances are called counterfactuals. Theories containing counterfactuals may be plausible, but they cannot be proven true.

Social scientists want to have theories capable of being generally true for as many marriages and families as possible. When exceptions are found, they can explore why the exceptions occur. Good theories also must be capable of disproof. If there is no way to disprove them, outrageous claims can be made, and there is no effective way to argue against them. Because a couple cannot turn back the clock and behave differently, it will never be known what caused their divorce.

Family scientists base their theories on information enabling comparisons across many cases. If many similar couples can be found, and all of them get divorced, scientists might be closer to a general understanding of the causes of divorce. Moreover, if many couples who are similar in all respects but a few are found, and those with the differing circumstances do not divorce, scientists might form an even more useful theory about divorce. Instead of relying on arguments about what did not happen, they compare different couples who have different experiences, some ending in divorce and some staying married.

In this way, family scientists have developed many theories to guide their thinking (Boss et al. 1993). These theories differ from each other in several fundamental ways.

Philosophies of Family Science

Theorizing can be based on different ideas about how a science works. Three primary approaches can be distinguished. A positivistic philosophy of family science makes several assumptions:

1. There is a real world of family life. This world is a natural one, and it operates according to a set of general principles. Truth is a matter of discovery.
2. The world of family life is ultimately knowable. Through careful study of how individual families work, scientists can increase their understanding of family life.
3. The best way to study families is by using standard methods useful in other domains of scientific inquiry. Reliable and valid evidence, or factual information, must be collected, based on observing families.
4. With increasing knowledge based on the facts, social scientists can intervene or assist others to intervene to improve family life.

This positivistic or optimistic approach was dominant throughout the twentieth century. It arose to help family studies gain stature as a scientific enterprise not unlike the other more established sciences. It also helped to distinguish scientific theories about the family from other kinds of contemplation, based on theological principles or other beliefs about the "appropriate" ways of being a family. Social science should be conducted in a spirit of free inquiry, without interference from governments or other nonscientific authorities.

On the other hand, the critical philosophy of family science starts with the idea that all beliefs and practices are political. Families, as well as the scholars who study them, are engaged in a struggle for domination and respect.

Historically, it can be observed that certain kinds of families and family members have been dominated and oppressed by those who are in more powerful positions in society. The less powerful usually have been females, persons of relatively low socioeconomic status, members of racial and ethnic minority groups, children, and sometimes elderly members of society. Any person or group different from the image of a "normal" family is considered to be abnormal or deviant by members of dominant groups. Those people in power not only control material resources but also intellectual resources, the way society thinks about families. Most members of the scientific "establishment," including family researchers and theorists, have been members of dominating social groups and categories.

The critical perspective challenges not only the content of positivistic theories, but also the assumptions upon which positivism rests. There is no natural world of family life to be discovered. Instead, what seems natural is the product of political forces, and of the domination of thinking and acting by some privileged families, family members, and family scientists. Truth is not a discovery, but a weapon. The proper goals of science are not the accumulation of facts and theories based on them, but instead are enlightenment and emancipation. Theories should be used to expose domination, and to assist the transformation of society and of science itself into more humane entities, resulting in a better world. Such a world will be one in which diversity of both lifestyles and modes of thinking will be equally respected and allowed to flourish. Some feminist theories and social conflict theories of family life rely on a critical philosophy of science (Farrington and Chertok 1993; Osmond 1987; Osmond and Thorne 1993).

A third philosophy of science influencing family theory is the interpretive approach. This view claims that all reality is a human construction. There is no objective truth about families, only a variety of subjective views that are developed through a dialogue with others in an effort to achieve a shared and workable understanding. Whatever is claimed to be known is tentative, always in process, and always just one point of view within a stream of alternative and evolving views. Whatever entity is called a family, the members of that entity are principally engaged in negotiating a sense of meaning, one that enables them to better understand who they are and how they fit into the environment.

Interpretive family theorists tend to reject positivism as naive, as making ideas seem firmer, more factual, or more stable than they really are. Truth is not a discovery, but an invention. The purpose of theorizing about the family is to make a personal statement based on an inevitably limited view. Instead of finding theories that will stand the test of time, the best theories are about the search for meaning in which families participate. These understandings and the processes by which they are created should be part of the content as well as the method used by family theorists. Interpretive theorists and researchers let family members speak and act for themselves and observe how reality is socially constructed. The theories of scholars then emerge and change as the theories created by families emerge and change. Symbolic interaction theory and phenomenological theory usually rely on an interpretive philosophy of science (Gubrium and Holstein 1993; LaRossa and Reitzes 1993).

All three philosophies influenced family scholars throughout the twentieth century. Because they are philosophies, there is no positivistic way of deciding which is best. The preferences of family theorists relate to the way they were trained, the acceptance by their colleagues of the alternatives, and the personal lives and other professional experiences of those in the scientific community (Klein and Jurich 1993; Thomas and Wilcox 1987).

These philosophies do not represent entirely incompatible viewpoints. Some family theorists accept the usefulness of more than one philosophy, even if they rely on only one of their own theories. Others combine features of two or more philosophies of science when they theorize.

Purposes of Family Theory

Theories about the family also differ in terms of the purposes that theorists have in formulating them. The most common goal is to provide a general description of how families work. In order to achieve a useful description, theories contain concepts. These concepts, such as cohesiveness, size, or patriarchy, help to compare families, and commonly have technical definitions. Family theorists usually strive for clear and precise definitions, so that they may measure what happens when families are directly observed or when members report their ideas, feelings, and behaviors. Many concepts are treated as variables, properties with different quantities on some scale. For example, families may be more or less cohesive, larger or smaller, and more or less patriarchal.

Different types of concepts are used in family theories. Some point to the structure of a family, its composition or the way it is organized. Some concepts describe patterns of social interaction, the quality of relationships, or processes that occur in families. Some theoretical concepts show how other concepts are related to each other. For example, if a family has five members and one descriptive concept refers to how flexible each member is, the family itself may be flexible if it meets a certain level of flexibility in its members. Perhaps all members must be at least halfway flexible, or perhaps some of the five must be very flexible to compensate for the inflexibility of the others. Whatever concepts are used, it is impossible to have a family theory unless there is a fairly detailed vocabulary for describing what makes families similar to and different from each other.

Most family theories go beyond description and provide an explanation. To explain something, it is essential to argue why it occurs. There are two basic ways to explain family life (Burr et al. 1979).

One type of explanation uses a deductive argument. This begins with a small number of premises (or axioms), statements individuals are willing to assume are true. Then, other statements (or theorems) are logically derived from the premises. Consider the following illustration:

1. All social systems are goal-directed. (Axiom 1)
2. All families are social systems. (Axiom 2)
3. One goal of all social systems is survival. (Axiom 3)
4. All families direct energy toward survival. (Theorem 1)

The theorem may be true, but only if all three axioms are true, in which case an explanation for why families direct energy toward survival exists; they do so because of the meanings inherent in the axioms. If the illustration were a real deductive explanation, additional information would have to be provided. The meaning of social system, goal, family, and energy would have to be defined, and more theorems would be derived. Deductive explanations are usually considered to be powerful if many theorems can be derived from a small set of axioms.

Notice how an explanation is achieved in this example. Families are treated as one type of social system, and survival is treated as one type of goal. Subsuming one phenomenon under a broader phenomenon is a common way of making a deductive argument. Another common way is to link statements in a chain. For example:

1. If parents encourage their children to explore the environment, children will explore the environment. (Axiom 4)
2. If children explore the environment, they will have high self-esteem. (Axiom 5)
3. If children have high self-esteem, they will effectively solve their problems later in life. (Axiom 6)
4. If parents encourage their children to explore the environment, children will effectively solve their problems later in life. (Theorem 2)

Here, effective problem solving for some people has been explained by referring to a chain of events that produces it. Furthermore, it has been argued that these axioms are transitive. That is, by having a series of statements with then in one becoming if in the next, it is possible to see a link between two ideas (in this case, what parents do and how children solve problems), a link that previously may have gone unnoticed. The argument may require elaboration before it is satisfactory, however. Simple if/then statements may hold only under special conditions. For instance, it may be argued that other things must be present, such as a willingness on the part of children to do what parents encourage them to do, before they will act as Axiom 4 argues.

While deductive explanations tend to be clear about the logic underlying an argument, family theorists have found them to be of limited value. The main reason is that it must be assumed that the premises are true. It often is difficult to make a convincing case that they are true. Different premises might be created that lead to the same conclusions, which would raise doubt about the original premises, or further research might prove that some theorems are false. Either of these situations would indicate that something is wrong with the original deductive explanation, but it would not pinpoint the problem.

The most popular way to explain family life uses a causal argument, which starts by assuming that everything that happens has some cause. The way families are or the actions they take cannot just be accidental. Some actions or conditions in the past exert influence on the current situation. An explanation is achieved by first showing how families are different from each other and then showing how differing prior circumstances are responsible for the differences to be explained.

In its simplest form, a causal argument is deterministic. It is assumed that there is one primary causal factor and it completely determines the result. In practice, however, family scholars have realized that causes are seldom so simple. Causal explanations generally employ several antecedent factors or conditions, working together or as alternatives, and all of them are included in the argument. Each causal element only works to increase the probability that a particular outcome will occur.

Causal explanations may show several factors converging to influence one outcome. They may show several separate paths, with several intervening steps, before an outcome is reached. They may even specify the conditions necessary before one variable can have an influence on another variable. In any case, causal explanations require fairly stable and strong connections between variables. Causal factors must happen chronologically before the effects occur, and the connections must not be just coincidental, byproducts of something else that is the "true" cause.

While causal explanations in family science have been popular, they are often viewed cautiously. Even the best ones tend to explain only a modest fraction of the differences between families. To improve them, there often is a temptation to make causal arguments more complex. If they grow too complex, however, they begin to lose their intuitive appeal. It is a challenge to understand what a very complex casual argument is really claiming. Part of the attraction of causal arguments in family theories is that the technology for using them to conduct empirical research is well developed. This technology involves statistical skills that sometimes seem remote from the family lives the researchers are trying to understand.

One problem with causal explanations is the frequent requirement that scientists follow families over time, because the families are supposed to change due to causal forces. Quite often, however, researchers compare different families at one time, and changes within families are not observed. Thus, researchers may be tempted to think that they have found causes, when they really have only found associations between variables.

Another problem with causal explanations is the mistaken belief that it is possible to explain what usually happens causally. Suppose it is discovered that 30 percent of all children in the United States are born to single mothers. Researchers may want to know the cause of this percentage. Then, they may identify a possible cause, perhaps the advantages of staying single. When this idea is included, the problem of counterfactuals is faced again. Single mothers may have common experiences suggesting the advantages of singlehood. But social scientists must compare single mothers with other mothers, and they must also compare all of the mothers with regard to the proposed cause. Family scientists can never causally explain what often or always happens in families or the average family experience. Instead, causal explanations can only explain differences, in this example why some mothers are married when they bear children while other mothers are not.

Some critics of causal explanations argue that the entire enterprise is misguided because scientists never can prove that something is a cause or part of a cause. Nevertheless, theories that rely on causal explanations remain popular in the family field, even if they cannot provide complete explanations. Causal explanations provide a useful way to think about family life, and the challenge is to make them better than rival causal explanations, not to assume that they ever will provide the final word or the perfect theory.

Most family theorists who provide descriptions and explanations believe these are the two most important purposes to achieve. Some family theorists, however, want to show how to change families by intervening to do something for their benefit. This goal is based on the idea that some families are not functioning well and that it is important to solve family problems or prevent them from occurring.

Interventions to change families must be based on an evaluation of the current situation and a decision that some families should be altered to reach an objective not now being met. Because people often disagree about goals and standards, any intervention relies on a point of view. Families themselves may determine that they are not meeting their own goals. A theorist may have goals for families that do not correspond with a family's own goals or values. The standard selected may be some notion of what is generally acceptable in society at large.

Whose goals should direct an intervention is often controversial. Consider the discovery that physical abuse of children by parents is fairly common in the United States. A scholar may develop a good theory about why some parents abuse their children and others do not. Perhaps one causal factor in the theory is the extent to which parents feel they have the right to punish children as they see fit. Those parents who feel that severe physical punishment is acceptable then use this form of punishment. A good theory should allow the theorist to determine what needs to be done to reduce the likelihood of child abuse. In this example, what is needed is a change in the belief by some parents that their behavior is acceptable. The problem is that parents may not feel that their punishing behaviors are unacceptable. The only way to avoid controversy surrounding the use of family theories to change families is to identify a goal that everyone accepts.

Even if there is no controversy over goals and values, it may be difficult to implement the desired change. If the theory implies that families must be changed, a program of action must be developed to reach families and change them. Sufficient confidence in the theory must exist so that a change in the causal factors has a good chance of producing the desired effect. This often requires careful research, because undesirable consequences of well-intentioned changes may occur. Finally, the required change in the cause may be difficult in principle to produce. If, for example, a theory argued that the basic fabric of society must be changed in order to reduce child abuse, figuring out how to change the fabric of society would be a tall order.

If a theory explains well what has happened in the past, it should provide a good prediction about the future. Another purpose of family theory is to enable accurate estimates of what families will be like in the future. Therefore, once a theory has been formulated, further research must be conducted to see if the theory remains useful. The connection between past and future, however, depends on a fairly stable environment. Some family theories do not survive events that take place after they have been formulated. This usually means that the original theory must be revised to reflect changes in families and in their environments more accurately. If a theory cannot be revised, it tends to be discarded.

Difficulty predicting family life may not be a serious deficit. The future is difficult to predict in many areas of science. Nevertheless, it is important to notice when a family theory was developed, and to find out what has happened subsequently. If an older theory about the family is encountered that no longer seems popular, newer literature can be examined to see if this loss of popularity is due to faulty prediction or a failure to revise the theory. Family theories usually are not static entities. They tend to change as families and their environments change, and as new theorists with new insights join the field.

Meaning of Family

Another important difference among family theories is in the way their central topic, the family, is defined and used. While all theories have a descriptive purpose, not all family theorists view families identically. In fact, they view families according to four different meanings of the term family.

One way to look at families is based on structural features. Families contain varying numbers of persons who are related in particular ways, including such persons as mothers, fathers, and children. This view may be extended to include grandparents, in-laws, step-relations, and perhaps even former relatives. Structural definitions of family focus on the composition of its membership. They may indicate that family members are related by blood, marriage, or some other legal bond such as adoption. Sharing a household may be another structural feature. With a structural definition, the theorist is able to determine which kinds of social groups do not qualify as families and which individuals are in a particular family.

Structural definitions of family also attend to the types of relationships that create social bonds between members. Important bonds are created by communication, power, and affection, as well as the daily work and leisure performed by family members. Scientists can observe how patterns of social interaction among the members are structured, and they can specify the various rules or principles that families use to organize their activities. Families may be structured by such characteristics as gender, age, and generation, as well as their connections to the outside world. These structures also are useful for distinguishing families from other kinds of social groups and organizations.

Theories about the family usually focus on some limited structural form. For example, they may apply only to married couples or to mothers and daughters. Sometimes theories compare different family structures. A theory might deal with how parent-child relations differ when two-parent families are compared to mother-led families.

A second way to look at families is based on functional elements. Why do families exist in the first place? Every human society has families, so they must serve some generally recognized purpose or function. Most functional definitions of the family focus on the importance of human reproduction and the necessity of nurturing dependent children for a relatively long period of time. Functional family theories often address the structural variety of families, with assertions about how effective each structure is in accomplishing the requisite functions that families everywhere have. From this perspective, if a certain structure does not fulfill some family function, families with that structure may be considered to be dysfunctional families.

A third meaning of family is based on interactional features, that is, it emphasizes repeatable processes of social interaction within families. Such interaction may be patterned or structured, but the focus is on the ongoing activity within the family, often conducted jointly by the members or otherwise coordinated. Family theories that rely on an interactional definition include concepts and variables describing what each participant is doing, how the members influence each other, and the quality of their relationships. From this perspective, a group need not have any particular structure to be counted as a family. Any social group that acts like a family would qualify as being a family. Social exchange theories often adopt an interactional view of family relationships (Sabatelli and Shehan 1993).

The fourth meaning of family is based on symbolic elements. Focus is on the meanings, perceptions, and interpretations that people have about family experiences. Only by watching how persons communicate or use dialogue to construct, challenge, and alter meanings do social scientists come to understand what a family is. Often this expression is verbal. The symbols people use to create and recreate family go beyond spoken words, however. Other important symbols are nonverbal intonations, bodily gestures, practices of dress and grooming, written statements, and visual images such as photographs and the spatial arrangement and condition of possessions in the home. Family theories based on the symbolic perspective emphasize various languages used to communicate, as well as the many artifacts with symbolic meaning created by families.

These four meanings of family are not always used separately. Two combinations are especially common. A combined structural and functional perspective informs structure-functional theory (Kingsbury and Scanzoni 1993). A combined interactional and symbolic perspective informs symbolic interaction theory (LaRossa and Reitzes 1993).

Each of the four meanings of family can be used alone, however. For example, it is possible to have a structural theory about some aspect of family life, perhaps offering structural causes of some limited family activity, without implying anything about the functionality of what is explained. For instance, the size of families or the size of communities in which they live might influence the amount of companionship among family members. It is also possible to use patterns of interaction as a cause or as the outcome in a family theory, without incorporating any ideas about the symbolic significance of the interaction to the family's members. For example, how often family members argue may influence how household chores are performed.

Level or Scope of Family Theories

Theories about the family differ in terms of their breadth of vision, level of analysis, and scope. Microscopic theories tend to focus on the internal workings of families, viewed as small groups of people in fairly intense relationships.

Mesoscopic theories focus on the transactions between families and people in the near environment who represent other groups and organizations. At this level, family theories are concerned with such things as friendships between members of different families, and the linkages between families and schools, churches, places of employment, the mass media, retail firms, and other public or private facilities and organizations.

Macroscopic theories concentrate on how the family as a social institution is embedded in society at large or in the nonhuman environment. They may, for instance, address how contemporary ways of family living emerged from significant changes in the economy, in national politics, or in technological developments. Structural and functional theories tend toward the macroscopic end of the spectrum, while symbolic and interactional theories tend toward the microscopic end.

Scope is a relative matter. For theorists of the human family, the social unit called family is roughly at the center of the spectrum, so that moving outward makes a particular theory more macro and moving inward makes it more micro.

Some family theorists focus on a fairly narrow range of the spectrum and formulate all of their ideas at one level or another. Other family theorists deliberately bridge levels, creating a transcopic theory. These multilevel theories often argue that phenomena at one level are the causes of phenomena at another level. Among such theories, the most common is a top-down approach. Societies affect families, and families in turn affect the individual persons in them. Increasingly popular are bottom-up theories that simply reverse the direction of causation, and reciprocating transcopic theories that emphasize mutual causation between levels in either alternating or simultaneous patterns. Family theories based on ecological principles currently are popular among those that are transcopic (Bubolz and Sontag 1993).

The scope of a theory helps scientists see the amount of causal agency attributed to families, as opposed to other factors outside or inside the family. Some theorists argue that families are primary causal agents. What they do has important consequences, and what makes them act may be important but is not addressed in the theory. Other theorists take exactly the opposite approach. Phenomena at the family level are to be explained by forces external or internal to them. If a theory remains entirely at the family level, it will explain something about family life in terms of causes elsewhere at the family level. A causal theory must have at least some cause or some effect at the family level to really be a theory about families. Some theories are called family theories even if they deal with only parts of a family, such as a theory about divorce or about the relationship between grandparents and grandchildren.

Time Perspectives

All theories about the family deal with the flow of chronological time, and sometimes with the social and psychological organization of time. Four principal time perspectives are common: static, episodic, biographical, and epochal.

In some theories, time is suspended or relegated to the margins. The idea is to craft a theory that is timeless. This static picture may be useful, especially if it is a general description. Given the previously noted problems associated with change, however, static family theories are themselves not very durable.

Often, the image of time is episodic. A process is being described and perhaps explained, and it is temporary. The entire process may last a few moments, a few days, or several months. If scientists trace what is happening over the course of events contained in the theory, everything can be observed with moderate effort when the theory is tested.

Another image of time is biographical. This perspective usually considers the entire span or course of life. Family phenomena begin at birth, develop through time, change along the way, and end when life ends. The idea of a "life" comes from the study of individual organisms, and it must be adjusted to speak meaningfully about the lifetime of a social group containing several organisms.

One adjustment is to consider the "birth" of the group to occur when the group itself forms, with the "death" of the group corresponding to the dissolution of the group. Some of the members will be alive before the group forms, after it dissolves, or both. New members may be added after the family forms, and some may be lost before the family ends. Families may endure even with great turnover in membership, as for example a lineage that survives over many generations. Individual persons may have experiences as members of several different families over the course of their own lives. Becoming a widow, getting divorced, remarrying, and giving birth to or adopting a child are among the events marking the course of both an individual's life and the life of the group.

Because it is usually impractical for one scholar to study large numbers of families from their formation to their dissolution, many theories that deal with biographical time concentrate on a shorter time segment. Some describe and explain what is happening during a particular stage of family life, such as when children are adolescents or after all of the children have become adults and left their parental homes. Another common alternative is to focus on a particular transition period. For instance, some theories focus on the process by which couples get married, or why some get married and others do not, tracing events from first meeting to the early years after marriage or until a breakup before marriage. Other time-limited biographical theories concern the transition to parenthood, the transition to the "empty nest," and so on. Family development is the most common name for the theories that treat families in biographical time (Rodgers and White 1993).

The other image of time is epochal. Fairly broad sweeps of history may be examined and categorized into periods. Families in ancient Greece, families in the American colonies, families during the early industrial era, and families during the Great Depression represent some of the historical categories that may give focus to a family theory. Other theories take a more sweeping historical perspective. Theorists may wish, for example, to explain how human families evolved from primate families, or how the contemporary family emerged from forces at work over several centuries. While many family theories using an epochal image of time are descriptive, evolutionary or biosocial theories of family life usually are explanatory as well (Troost and Filsinger 1993).

Forms of Expressing Theory

One useful way to differentiate theories about the family concerns the way they are expressed by their authors. Some theories are written in narrative form. They use prose expressed in commonly understood language. Other family theories are somewhat more formalized and are called propositional. A theorist identifies a set of well-bounded, declarative statements that serve as the theory's core propositions. Many of the concepts in these statements have technical meanings, and definitions are included. Often, the propositions assert how two or more variables are related, how strong the connections are and when they happen, and whether or not causal influence is implied. Theories that use shorthand, technical expressions are even more formalized. They contain mathematical symbols, diagrams with arrows, flow charts, or figures with classifications into types.

All forms of expression have virtues and limitations. More formalized theories are precise, and they are easy to distinguish from other theories with similar content. If a theory is imprecise or fuzzy, it is difficult for the scientific community to agree on what is meant, and extremely difficult to demonstrate that some of the arguments may be incorrect. Formalized theories require specialized training to be fully interpreted, however. Because technical expressions are arbitrary and may require intricate rules, some family theorists avoid them. Some avoid highly formalized theories because they can dehumanize the subject matter and place more emphasis on the structure of an argument than on its content. A truly good theory may be one that either combines forms of expression or can be translated from one form to another without changing its meaning.

Methods of Creating Theories

Theories about families usually develop over time as theorists incorporate prior knowledge and new experience. At first, there may be only fragments, enough of an argument to share the basic shape of a theory with an audience. If a particular theory has been discussed for a period of time and a consensus has been established, the theory may be named and only brief mention made of its details, on the assumption that colleagues understand what is involved. Working to produce a family theory, however, usually takes place in one of two ways.

Deductive theory is produced by starting with fairly abstract ideas and without particular regard for the way families can be observed to operate in the "real" world. Some of the ideas may be borrowed from other areas of study, and some may represent the integration or modification of existing ideas about family. Portions may be entirely new, but more often the theorist is just reshaping or recombining ideas that have appeared in other scholarly works. Theorists may work deductively even when they are not creating deductive explanations.

Once the new theory is given a clear structure, the theorist or a colleague who is attracted to the theory conducts empirical research to test some of the arguments. If the theory is supported by research data, gathered and analyzed using suitable methods, the theory is provisionally accepted. This acceptance is provisional because it takes repeated tests, often by different groups of researchers using somewhat different methods, before a great deal of confidence in the theory is warranted. If the theory is unsupported or refuted by research data, it is revised or discarded in favor of a superior alternative. Ideally, two rival theories with different explanations and predictions are pitted against each other in a single study or a carefully managed series of studies. This enables scholars to determine which of the two theories is better.

Some family scientists object to the deductive process. While they acknowledge that it is the usual textbook approach, they offer either of two arguments. The weak theory objection is that scholars really do not use the deductive method. Instead, they are guided by hunches derived from the direct experiences they have, either as handlers of empirical data or as participants in family life. The strong theory objection is that every judgment and decision a scholar makes is based on preconceived ideas to which that scholar has very strong attachments. Because all social scientists have ideas and beliefs about families, the theories they create are biased in ways that escape the attention of even the most impartial theorist.

To take advantage of both objections, some family scientists use an inductive method to create their theories. In its pure form, the scientist disregards all previous knowledge and speculation about the topic of interest. Research with minimal biases is conducted, and the participating families and the results they produce are taken at face value. A useful theory is developed either after the research is concluded or slowly during the process of study. A grounded theory emerges.

Much family theorizing is transductive, with elements of both deduction and induction. The two extreme approaches provide models for how to work, but there is room for an intermediate approach.

Many participants are involved in the process of creating any theory. Even if only one author receives credit, that person builds on the ideas of others. If a particular theory has many acknowledged contributors and if it endures sufficiently long, it becomes recognized as a theoretical tradition or school of thought. The family members who participate in the creation of family theory may be recognized as coauthors, but often they are not.

Other Differences

Family theories can be distinguished in additional ways. Some theories are relatively abstract and speculative, while others are more concrete and stated in language closer to observable phenomena.

Some family theories are quite general, while others are much more context-specific. General theories are claimed to hold regardless of time or place, or apply to broadly encompassed times and places. Context-specific theories tend to focus on restricted populations, such as one culture or society, the families in one social class, a segment of families with a narrow age structure, one gender, or one racial or ethnic group. Some family theories entail comparisons across contexts, but without covering all of the possibilities. The context of time also varies between family theories. Whether they adopt episodic, biographical, or epochal images of time, most family theories concerned with processes of change carve out a limited span of time for their arguments.

Theories about the family also differ in terms of the breadth of content they cover and which particular subunits within the family are addressed. Theories may be narrow, middle-range, or broad in their content. Relatively speaking, a theory about the effectiveness of communication by husbands is narrow, while a theory of marital quality is middle-range, and a theory of family functioning is broad. In this example, not only does the subject matter become broader with the move from narrow to broad theories, but the relevant units also become broader.

Family theories differ considerably in complexity. Simple theories may involve no more than two or three concepts and two or three relationships among them. Complex theories contain a large number of concepts or variables, and many links exist among the concepts.

Finally, family theories differ according to how coherent a picture of family life they present. Some theories represent families as fairly atomistic collections of elements; scientists understand families if they understand how their elements work, and understanding is impeded if elements are combined that really do not go together. Other theories are more holistic, because they focus on the family as a totality; while families may have elements or features, scientists do not understand how families work unless they see how the features are connected and how these connections produce something unique. Family systems theory is an example of a popular holistic theory (Broderick 1993; Whitechurch and Constantine 1993).

The ways family theories differ in their abstractness, generality, breadth, complexity, and the coherence of their imagery are all matters of degree. While there is much diversity, there also are unifying efforts. Abstract theories can be made more concrete, context-specific theories can be made more general, and complex theories can be simplified, among other possibilities. Part of the ongoing excitement about theorizing in the family field is that there is an endless array of projects for enterprising theorists.

Conclusion

There are several reasons for the diversity among family theories. One is the growing number of scholars who have taken family life as an area of serious investigation and the rapidly expanding library of works they have produced. Family theorists also represent a large number of academic and applied disciplines, in and beyond the social sciences. Their ideas are shaped by the specialized training they receive and the different missions established in each discipline. Finally, some questions seem to be answered more satisfactorily if one type of theory is used instead of another.

As long as families remain a central domain in the way people think about the world, and as long as family life is sometimes viewed as troubled or problematic, there will be a sense of urgency about increasing understanding of families. The result is predictable: more theories, more research, and more programs proposed to change what can be changed and to accept what cannot.

See also:Dialectical Theory; Family Development Theory; Family Systems Theory; Family, Definition of; Family, History of; Human Ecology Theory; Kinship; Life Course Theory; Phenomenology; Role Theory; Social Exchange Theory; Structural-Functional Theory; Symbolic Interactionism

Bibliography

boss, p. g.; doherty, w. j.; larossa, r; schumm, w. r.; and steinmetz, s. k., eds. (1993). sourcebook of family theories and methods: a contextual approach. new york: plenum.

broderick, c. b. (1993). understanding family process:basics of family systems theory. newbury park, ca: sage publications.

bubolz, m. m., and sontag, m. s. (1993). "human ecologytheory." in sourcebook of family theories and methods: a contextual approach, ed. p. g. boss, w. j. doherty, r. larossa, w. r. schumm, and s. k. steinmetz. new york: plenum.

burr, w. r.; hill, r.; nye, f. i.; and reiss, i. l., eds. (1979). contemporary theories about the family, vol. 1: research-based theories. new york: free press.

farrington, k., and chertok, e. (l993)."social conflict theories of the family." in sourcebook of family theories and methods: a contextual approach, ed. p. g. boss, w. j. doherty, r. larossa, w. r. schumm, and s. k. steinmetz. new york: plenum.

gubrium, j. f., and holstein, j. a. (1993). "phenomenology, ethnomethodology, and family discourse." in sourcebook of family theories and methods: a contextual approach, ed. p. g. boss, w. j. doherty, r. larossa, w. r. schumm, and s. k. steinmetz. new york: plenum.

kingsbury, n., and scanzoni, j. (1993). "structural-functionalism." in sourcebook of family theories and methods: a contextual approach, ed. p. g. boss, w. j. doherty, r. larossa, w. r. schumm, and s. k. steinmetz. new york: plenum.

klein, d. m., and jurich, j. a. (1993). "metatheory andfamily studies." in sourcebook of family theories and methods: a contextual approach, ed. p. g. boss, w. j. doherty, r. larossa, w. r. schumm, and s. k. steinmetz. new york: plenum.

larossa, r., and reitzes, d. c. (1993). "symbolic interactionism and family studies." in sourcebook of family theories and methods: a contextual approach, ed.p. g. boss, w. j. doherty, r. larossa, w. r. schumm, and s. k. steinmetz. new york: plenum.

osmond, m. w. (1987). "radical-critical theories." inhandbook of marriage and the family, ed. m. b. sussman and s. k. steinmetz. new york: plenum.

osmond, m. w., and thorne, b. (1993). "feminist theories: the social construction of gender in families and society." in sourcebook of family theories and methods: a contextual approach, ed. p. g. boss, w. j. doherty, r. larossa, w. r. schumm, and s. k. steinmetz. new york: plenum.

rodgers, r. h., and white, j. m. (1993). "family development theory." in sourcebook of family theories and methods: a contextual approach, ed. p. g. boss, w. j. doherty, r. larossa, w. r. schumm, and s. k. steinmetz. new york: plenum.

sabatelli, r. m., and shehan, c. l. (1993). "exchange andresource theories." in sourcebook of family theories and methods: a contextual approach, ed. p. g. boss, w. j. doherty, r. larossa, w. r. schumm, and s. k. steinmetz. new york: plenum.

sprey, j., ed. (1990). fashioning family theory: new approaches. newbury park, ca: sage publications.

thomas, d. l., and wilcox, j. e. (1987). "the rise of family theory: a historical and critical analysis." in handbook of marriage and the family, ed. m. b. sussman and s. k. steinmetz. new york: plenum.

troost, k. m., and filsinger, e. (1993). "emerging biosocialperspectives on the family." in sourcebook of family theories and methods: a contextual approach, ed.p. g. boss, w. j. doherty, r. larossa, w. r. schumm, and s. k. steinmetz. new york: plenum.

whitechurch, g. g., and constantine, l. l. (1993). "systems theory." in sourcebook of family theories and methods: a contextual approach, ed. p. g. boss, w. j. doherty, r. larossa, w. r. schumm, and s. k. steinmetz. new york: plenum.

david m. klein (1995)