Close this search box.

What is Bowen Theory?


This rather lengthy and academic article is taken from the dissertation section “Bowen Theory” in the chapter “OverView of Bowen Theory and Systems Philosophy here: It is not a current and refreshed view of the author, but is posted here to serve as at least some introductory survey of Bowen theory while a better introductory survey can be compiled. An improved survey will likely use the concept of anxiety as an entry point, and proceed through the concepts to explain increasing amounts of variation in the human individual and family phenomena.

Bowen Theory

In a seminal publication co-authored with Murray Bowen, Michael Kerr (M. Kerr & Bowen, 1988) described the shortsighted nature of our approach to the most important human problems, “We demonstrate against war as if we understand the causes of war. We could just as easily demonstrate against schizophrenia” (p. 27). This critical view of what is known and what is not known, and how our inability to distinguish between assumption and fact, will be an important emphasis if this section on Bowen’s way of thinking and the theory that is a product of that way of thinking.

I have found BT to be a tightly integrated theoretical system of interrelated concepts which define predictable patterns of human behavior. It is not possible to accurately portray the theory and the type of thinking that it represents with the space provided here, nor is it necessary. Bowen defined a natural system theory of the human family as an emotional unit, and Gotama did not define a theory of the human family per se. Therefore, this section will only outline important aspects of Bowen’s work and style of thinking which pertain to the research question, “To what extent did the Buddha define a natural system?” Other concepts will only be briefly explained. 

Bowen and Systems Thinking

Murray Bowen was unique in his theoretical research on human behavior from a natural system perspective. There are many systems theorists who use models of human behavior derived from general systems or cybernetics ideas, such as Minuchin, Jay Haley, and Gregory Bateson. These efforts were mostly clinical in nature and focus more on deducing clinical intervention from existing theory than inducing new theory from naturalistic observation. There are many researchers who study a single aspect of human behavior from a biological perspective (Stanford, 2011). One common trend among these researchers is that they pass from the cellular, organic, mind–body levels and jump over the family to the social level (Noone & Papero, 2015b). The literature reviewed here reveals no other systems theorists who study the human family as a naturally selected unit which governs its functioning through concurrently reciprocal processes among the individuals in the manner of a complex system. This conceptualization of the family as an emotional unit was Bowen’s primary theoretical contribution to the study of human behavior (Bowen, 1978).

Perhaps more important than the concept of the family as an emotional unit was Bowen’s approach to scientific research, which operated on the assumption of biological phenomena, in general, as emotional systems. This view provided for a unique research methodology, which could potentially move beyond reductionistic statistical models. At age fifteen, when he was working as an “ambulance helper” (Bowen, 1978, p. 483), he witnessed “bewildered, unsure, and fumbling” (Bowen, 1978, p. 483) emergency medical personnel fail to coordinate care for a dying girl, and from that point he thought he could help medicine find better answers. As with many of the doctors returning from the Second World War, he was interested in improving the treatment of psychological problems. He trained as a psychiatrist and eventually became chief resident at the Menninger Clinic in Topeka, Kansas, which was the foremost psychoanalytic clinic in North America at the time. During his time at Menninger, he began developing new ideas about the influence of family members on the symptoms of his patients. As discussed previously, he understood that Freud had intended to create a science of human behavior but began to suspect that the assumptions of psychoanalysis were too deeply rooted in human subjectivity. He had great respect and admiration for Freud as a “genius” theoretician with a unique ability to remain objective while in contact with his patients’ great emotional distress (M. Kerr & Bowen, 1988, p. 352), and recognized Freud as contributing one of the most important advances in the understanding of human behavior since Darwin.

But Bowen believed that Freud’s fatal mistake was straying from bare observation of nature by using concepts which could not be directly connected back to the accepted science. He writes that Freud had created a theory that was “unwittingly tinged with feelings when it was based on the history of human civilization rather than science itself” (M. Kerr & Bowen, 1988, p. 357). In his book Family Therapy in Clinical Practice, Bowen (1978) describes his opinion on the limitations of psychoanalysis. He writes,

Few events in history have influenced man’s thinking more than psychoanalysis. This new knowledge about human behavior was gradually incorporated into psychiatry, psychology, sociology, anthropology, and the other professional disciplines that deal with human behavior, and into poetry, novels, plays, and other artistic works. Psychoanalytic concepts came to be regarded as basic truths. Along with the acceptance there were some long-term complications in the integration of psychoanalysis with other knowledge. Freud had been trained as a neurologist. He was clear that he was operating with theoretical assumptions, and that his concepts had no logical connection with medicine or the accepted sciences. His concept of “psycho” pathology, patterned after medicine, left us with a conceptual dilemma not yet resolved. He searched for a conceptual connection with medicine, but never found it. Meanwhile, he used inconsistent models to conceptualize his other findings. His broad knowledge of literature and the arts served as other models. A striking example was the oedipal conflict, which came from literature. His models accurately portrayed his clinical observations and represented a microcosm of human nature; nonetheless, his theoretical concepts came from discrepant sources. This made it difficult for his successors to think in concepts synonymous with medicine or the accepted sciences. In essence, he conceptualized a revolutionary new body of knowledge about human functioning that came to exist in its own compartment, without logical connection with medicine or any of the accepted sciences. The knowledge was popularized by the social sciences and the artistic world, but few of the concepts found their way into the more basic sciences. This further separated psychoanalysis from the sciences. (pp. 338–339)

Bowen believed that “The use of concepts from literature separated [Freud’s] theory from facts that could be proven and validated by science” (M. Kerr & Bowen, 1988, p. 357). He believed that the psychoanalysts and psychiatrists who came after Freud had convinced themselves that the field represented a science because it used the “scientific method” (p. 391), which is a term he used to describe the application of statistics to subjective content which remains a cornerstone of research psychology today,

The twentieth century has been involved in a debate about whether psychoanalysis is a science. It is a science in the sense that it defines a body of facts about human functioning never previously described. It is not a science in the sense that it has never been able to make contact with, nor be accepted by the known sciences. The use of the scientific method has lulled psychoanalysis and psychiatry into believing it can someday become a science. The scientific method is a way of ordering random and discrepant data in a scientific way in the search for common denominators and scientific fact. Researchers have spent decades studying and restudying facts within psychoanalysis, discovering some new bits of information within the closed compartment, but they have not been able to make contact with the accepted sciences. Use of the scientific method does not make a body of knowledge into a science. (M. Kerr & Bowen, 1988, p. 391)

Bowen had a particular interest in solving difficult problems. He began reading extensively in biology, evolution, and the natural sciences to understand how the hard sciences had tackled new and difficult problems. He decided that any science of human behavior would have to be consistent with evolutionary theory, and he began looking for an institution which would support research based on his developing hypothesis on the predictable influence of interpersonal relationships on schizophrenia (M. Kerr & Bowen, 1988). He eventually moved to the newly created NIMH in 1954 and began the original study on the family where he would develop his natural system theory (Bowen, 1978, 2015).

Systems sciences were spontaneously emerging in different locations in the West and in various forms during the 1940s and early 1950s (Bertalanffy, 1968/2015). Laszlo (1971) and Bertalanffy (1968/2015) were developing systems philosophy and general systems concepts using mathematical models of nature. Wiener (1961/1985) was developing his cybernetic theory based on the idea of feedback loops as natural phenomena. The exact extent of the influence on Bowen by the emerging systems philosophy is unknown, though he claims that these emerging systems philosophies did not contribute to his thinking. But, perhaps his most important contribution was his effort to leave aside a priori assumptions to stick to a more basic form of scientific thinking. Bowen (1978) writes,

There is another common misconception that should be mentioned. Many believe that family systems theory, as I have developed it, came from general systems theory. That is totally inaccurate. I knew nothing about general systems theory when I started my research. It is a way of “thinking about thinking” which occupies the same position to divergent theories that the scientific method occupies in relation to divergent and discrepant facts. In the 1940s I attended one lecture by von Bertalanffy of which I remembered nothing, and one lecture by Norbert Wiener of which I remembered very little. Whether anything from those lectures found its way into my thinking is a matter for conjecture. I did extensive reading in biology, evolution, and the national sciences, which I believe led to my formulation of emotional systems theory on the model of “systems” in nature. (p. 398)

This passage implies that Bowen’s natural systems view was unique in origin. While it can be argued that Bowen took the term “system” from some element of academic culture at the time, he developed a simpler and more concrete meaning of the term which guided his research based on falsifiable prediction. This research methodology was integrated with psychotherapy as part of the experimental process, which in turn produced new observations that fed back into the research. The theory then emerged from observations based on his own notion of systems thinking. If the concepts in theory have any basis in the facts of nature, then it might be assumed that anyone thinking about and observing nature in this way would eventually discover them. Thinking in this way about Homo sapiens as a natural system is perhaps his most important contribution. Further, this move toward science as prediction by way of theoretical research as opposed to merely control by way of engineering is perhaps the most important aspect of BT of this study.

Michael Kerr, a close associate of Bowen, defines systems thinking as a broad category of science which focuses on processes and relationships instead of static, categorical properties (M. Kerr & Bowen, 1988). He defines systems thinking as a move toward nature and away from assumption and dogma. M. Kerr writes, if one equates “systems thinking with the ability to be aware of the process of nature instead of the content of nature, then there is evidence that systems thinking [in the West] dates back at least 2,000 years” (p. 14). M. Kerr reviews how the Greeks living in Ionia believed that the world was made of atoms and that everything that occurred now was the result of conditions propagated from the past, including their own existence.

M. Kerr (M. Kerr & Bowen, 1988) then associates Ptolemy’s geocentric theory with a sort of “pre-Ionian” regression for almost two millennia by claiming the following: “This conceptualization prevailed over the ideas of the Ionians and influenced thinking for more than 1,700 years! In addition, man continued to believe that he was created in his present form and that, yes, diseases were caused by demons” (p. 15). One marker for this type of thinking would be a reliance on overly simplistic, linear cause-and-effect thinking about a problem and the loss of the broader notion that causes are also effects. This linear “cause-and-effect” thinking is equivalent to what was previously discussed in the present study as linear thinking, or linear-causal thinking.

M. Kerr views Copernicus’ heliocentric model of the solar system, followed by Kepler’s predictive theory of planetary orbits, and Newton’s theory of universal gravitation as examples of developments rooted in systems thinking because they model processes instead of things. He argues that it is the fixation on the content of nature and lack of attention to the processes of nature that sets science back. He makes the case that Bowen’s theory of the human family as an emotional system draws its therapeutic efficacy precisely from developing the ability to “think systems” in this way (M. Kerr & Bowen, 1988, p. 158). M. Kerr distinguishes the natural system in the context of BT,

Rather than applying general systems concepts to the family, Bowen assumed that the family was a naturally occurring system. The word “natural” refers to something that pertains to nature, to something formed by nature without human intervention. The concept of a natural system, in other words, assumes that systems exist in nature independently of man’s creating them. The existence of natural systems does not even depend on the human’s being aware of them. The principles that govern a natural system are written in nature and not created by the human brain. The solar system, the ant colony, the tides, the cell, the family of homo erectus, are all natural systems. The human family system sprung from the evolutionary process and not from the human brain. We did not create it. We did not design human relationships anymore than the elephant or gibbon designed their family relationships. Family systems theory assumes that the principles that govern such things are there in nature for us to discover. (M. Kerr & Bowen, 1988, p. 24)

Caskie (1994) describes Darwin’s theory as an exemplary natural system theory, “Darwin’s theory of evolution was a natural systems theory that saw nature as a system, organized according to reciprocal relationships, mutually interdependent and mutually influencing. His theory described and proposed a mechanism which connected components of living systems” (p. 13).

Like classical psychological theorists, both Wiener and Bertalanffy developed systems views that assumed human behavior is more a function of what is unique to humans than what humans have in common with the rest of nature. Caskie (1994) writes that Wiener’s “first goal in building a theory was to smooth out the differences between the mechanistic and vitalistic positions in science through the use of common concepts derived from the field of communications engineering” (p. 9). He believed that man’s social structure was merely analogous to social structures in other species but that the faculty for language was the decisive organizer for human society. Similarly, Bertalanffy was interested in extending general systems concepts to human behavior and the rest of the sciences in an effort to unite the scientific disciplines. It can be inferred that he was more interested in man’s uniqueness and believed that the primary organizer for that uniqueness was “man’s capacity for creating symbols, and thus, culture” (p. 11). 

Bowen stands apart from system philosophers and theorists as one who prioritized the study of the Homo sapiensspecies as it is in nature by using concepts that only described what he observed in research. He used the kind of  inductive approach describe by E. O. Wilson (1998) for the development of theory. This approach began with as few assumptions as possible prior to making and interpreting observations. The resulting theory rests on the discipline through which he tested and refined his hypotheses to predict the onset and remission of processes at the group level. One example of the kind of factual, systems-level prediction that he made was that one process might begin in response to another process ending. This class of qualitative, process-level, complex-systems prediction is differentiated from predictions of what content will occur within a process. An example of a content-oriented prediction would pertain to precisely what a person will say and when as opposed to the fact that they simply said something (Bowen, 1978).

M. Kerr writes that, just as Johannes Kepler’s work on discovering that the orbits of the planets were elliptical instead of circular, a hypothesis has to be refined if it does not explain any single observation. He claims,

It has always been the task of science to modify theories and models to fit observations as opposed to modifying or ignoring observations to preserve existing theories. Kepler, although often frustrated by the existence of observations that did not quite fit his models, persisted until he was finally rewarded with a mathematically precise model that accurately described all the motions of all the planets. (M. Kerr & Bowen, 1988, p. 16)

This ideal became the rule in the psychiatric ward during Bowen’s original NIMH family study. The entire staff was involved in contributing to theory as a part of therapeutic work in the clinical ward (Bowen, 2015). This “research attitude” created an air of curiosity and objectivity among the staff which also played a key part in the application of new theoretical concepts and terms to clinical problems, and most importantly served as a model for patients to adopt themselves (Rakow, 2016). Bowen (1978) detailed his approach to science and his uniquely disciplined approach to clinical research during the NIMH study:

Psychotherapeutic principles and techniques were developed for each clinical situation. The hypothesis also predicted the changes that would occur with the psychotherapy. When research observations were not consistent with the hypothesis, the hypothesis was modified to fit the new facts, the psychotherapy was modified to fit the hypothesis, and new predictions were made about the results of the psychotherapy. When an unexpected clinical crisis arose, it was handled on an interim “clinical judgment” basis, but the hypothesis was considered at fault for not “knowing” about the situation ahead of time, and not having a predetermined therapeutic principle. The therapy was never changed to fit the situation except in emergencies. The goal was to change the hypothesis to account for the unexpected crisis, to change the therapy to fit the hypothesis, and to make new predictions about the therapy. Any failure to change in psychotherapy was as much a reason to reexamine and change the hypothesis as any other unpredicted change. Strict adherence to this principle resulted in a theoretical-therapeutic system that was developed as an integrated unit, with psychotherapy determined by the theory. (p. 520)

Bowen’s (1978) findings in this original study suggested that individual behavior is relatively unpredictable when conceptually separated from the group context, in particular the context of the family emotional system. Studying the family “as an emotional unit” (p. 192) suggests that the family operates as a single, multigenerational organism, and that variables which pertain to automatic processes in the group may account for variance on a higher order than variables which pertain to an individual. This implies a different conceptualization of the family or group as a single unit of analysis which makes it possible to make a new order of observations. Importantly, Bowen called the family unit a “system” in that “a change in one part of the system is followed by compensatory change in other parts of the system” (p. 179). If there is a change in a symptom in one person, theory predicts chances are good that it would be related to a functional shift in one or more parts of the patient’s emotional system.

The important difference in this research is the use of a different theoretical conceptualization of what is being observed. This reveals one axiom in BT that is often underemphasized in modes of family therapy, which the primary role that the way a person conceptualizes a problem plays in the action they take to solve it. These differences are apparent in the various modes of family therapy which emerged during the time of Bowen’s NIMH study. M. Kerr and Bowen (1988) write,

The way a therapist thinks about what energizes or drives the process he observes in a family will govern what he addresses in therapy. Many family therapists, for example, talk about the family being a “system” but they have many different ideas about what makes the family a system. (p. 11)

As discussed, Bowen’s approach to theoretical development is distinguished from other family researchers by only inducing concepts from research observations as opposed to the more common practice of applying pre-existing concepts from other kinds of systems. For example, Gregory Bateson, one of the more prominent family researchers of the time, used pre-existing machine-control ideas from cybernetics to conceptualize the family system (Bateson, 1987). Bateson might simply conceptualize relationship as a set of reciprocal transactions between two people in isolation, similar to attachment theory. His double-bind hypothesis is based on observations on a taxonomy of semantic abstractions used between mother and child which hold the child in a kind of emotional servitude with the mother. Nichols, a modern author of family therapy texts, might think of the family system as “an encounter between distinct interpersonal cultures” (Nichols, 2017, p. xxi). For Nichols, family therapy provides a mirror for one person to reflect their emotional experience onto another, similar to the classical encounter group concept. In that case, the goal of therapy might be to fill the room with an appropriate number of people to serve that purpose (Nichols, 2017). This appears to reflect something like superimposing individual concepts onto the family or group to achieve desired clinical outcomes rather than simply creating new conceptualizations to describe what is observed at the group level (Bowen, 1988).

One important distinction between Bowen and his fellow family therapy pioneers was that his peers were primarily interested in developing new therapies while he was interested in developing a scientific theory of human behavior. Bowen thought that prioritizing therapy over theory imposed preexisting assumptions onto the human as a natural phenomenon which might not be grounded in scientific theory. This is apparent in many theories of human behavior which impose pre-existing concepts from other fields without iteratively verifying and refining them through falsifiable predictions. The more this occurs, the more the theory is geared toward merely controlling the phenomenon prior to developing a sufficient understanding of how it works. Bowen’s approach was unique in that it was aimed first at prediction and second at intervention, even though the outcomes of intervention were fed back in as experimental observations in the NIMH study (Bowen, 1978).

M. Kerr and Bowen (1988) expand on this point, describing how the application of the term “system” to the human family as a product of nature differed from other family therapists in this subtle way,

Many biological and social theorists, for example, are convinced that the parts (cells, people, other organisms, or whatever they happen to be studying) so mutually influence one another that there exist “wholes” (body, family, or whatever) that must be understood as entities in their own right. The concept of “whole” implies that there exists an entity with principles of operation that regulate the functioning of the entity’s parts. A problem with the ideas of many of these theorists, however, is that they do not include a description of how the parts affect one another to create this “whole.” Without at least some idea about the “how,” it is quite easy to drift away from the realm of science and into the realm of holistic philosophy. . . . Systems is a descriptive term. It does not account for what is occurring, for what “drives” the process. . . . Saying that people function in reciprocal relationship to one another is a description of a phenomenon, not an explanation. . . . Saying that the human relationship process is rooted in instincts, has much in common with what occurs in other forms of life, and has a function in evolutionary terms is a step toward accounting for what occurs. This way of thinking about what “energizes” the phenomenon being described is contained in the concept of the family emotional system. (pp. 10–11)

Notably, M. Kerr’s metaperspective on the term “system” echoes Bowen’s critique of Freud’s and other family therapists’ use of “discrepant models” to develop theory which describes a single phenomenon. For example, Salvador Minuchin used the terms power, hierarchy, subsystem, boundary, alignment, coalition, triangulation, and others to describe different organizational structures which the therapist would seek to adjust in a family (Minuchin, 1974). But the emphasis appears to be on describing discrepant configurations within the family structure and not explaining the relationships between those structures and the predictable processes that they exhibit. In other words, Minuchin’s theory appears to be a categorical model where each configuration exists more within its own conceptual compartment. Without going into a detailed study of outcomes in structural family therapy, it may be possible that this conceptual fragmentation likely generates clinical interventions that are similarly discoordinated.

Because the relationships between the structures are loose, the terms themselves do not define a system where a change in one configuration directly predicts a change in another configuration. Thus, the application of each term to a particular case is left up to the therapist’s intuition, which would likely have served Minuchin as a remarkably intuitive master clinician. Therefore, theories like Minuchin’s designed more for control than prediction might serve more as a mental framework for organizing clinical interventions than the study of the family as a system which arises from nature. This, perhaps, suggests a distinction between clinical approaches based more on increased understanding where the outcomes vary less between clinicians, for example, in the treatment of a broken leg versus the treatment of an autoimmune disorder, where the outcomes may depend more on the talent of the physician than the science behind the medical practice.

M. Kerr, a psychiatrist, responds to this “erosion” of the terms theory and science in psychiatry as a postpositive science, arguing that “The trend of the decades had been one in which therapists interpreted theory according to their own feeling states” (M. Kerr & Bowen, 1988, p. 365). Bowen believed that human behavior would only be accepted as a science if it were “anchored in biology, evolutionary theory, and other knowledge about natural processes” (M. Kerr & Bowen, 1988, p. 5). Bowen thought that “systems thinking would provide the conceptual bridge from psychiatry to the accepted sciences” (M. Kerr & Bowen, 1988, p. 6). He believed that the “physical structure of the human brain was scientific, that the human brain functions to create feelings and subjective states, and that the brain is capable of separating structure from function” (M. Kerr & Bowen, 1988, pp. 354–355). Even the use of mathematics was in question as a conceptual framework that relied on ideas created outside the observation of a specific natural context and then often imposed upon nature to form a theory. “To get beyond mathematics and technology, I fashioned a natural systems theory, designed to fit precisely with the principles of evolution and the human as an evolutionary being” (M. Kerr & Bowen, 1988, p. 360).

Bowen broke from the mainstream in two ways: through the assumption that human behavior is mostly a function of factors universal to all of life, and the subsequent assumption that human behavior may appear unpredictable at the level of the individual but predictable at the level of the emotional system (M. Kerr & Bowen, 1988). BT contains eight concepts: nuclear family emotional system, DoS, triangulation, cutoff, family projection process, multigenerational transmission process, sibling position, and emotional process in society (Gilbert, 2006). M. Kerr writes, “None of the concepts was borrowed from psychological theory” (M. Kerr & Bowen, 1988, p. 13), explaining that theory development occurred as much in a vacuum as possible to allow for an entirely new way of looking at the individual. The orientation was instead a natural system theory and the focus was the shift from the individual to the family as an emotional unit. According to M. Kerr, this shift encourages one to “focus simultaneously on thinking/feeling behavior affecting atmosphere, and equal emphasis on atmosphere affecting each’s thinking/feeling behavior” (p. 9). According to M. Kerr, shifting the unit of analysis from the individual to the system and back again as often as the session demands requires a “quantum leap in the conceptual capacity of the observer” (M. Kerr, 1981b, 5:00).

The Emotional System

As described previously, Bowen’s primary assumptions were organized in the context of the facts of evolution, including Darwin’s theory of evolution. Facts of evolution describes observation of evolution itself, for example that there are many animals with a head, four limbs, and some kind of a tailbone, which is different than the theory of natural selection that describes those facts of evolution (M, Kerr, 1992). Definitions of research terms were assumed to come from or relate to biology. What resulted was not only a broad concept to organize research on human behavior, but a concept to simultaneously organize research on the individual, relationship system, and similar systems and subsystems in other species as well. Frost (2014) describes the emotional system as “behavior governed by the part of the human we share with the rest of life” (p. 304). The emotional system “describes the automatic processes by which an organism directs its response to the challenges and opportunities it faces” (Papero, 2015, p. 17). Kott (2012) writes that “What differentiates Bowen theory from other family systems approaches is its emphasis on the sensitivity human beings have to each other at a biological level” (p. 76). According to Bowen (1978), the emotional system handles the “myriads of sensory stimuli from the digestive, circulatory, respiratory, and all the other organ systems within the body as well as stimuli from the sensing organs that perceive the environment and relationships with others” (p. 372). In a detailed description on mammalian evolution, Bowen (1988) later writes, “The neocortex is designed for solutions of situations that arise in the external world. It receives signals primarily from the eyes, ears, and body wall” (p. 36). Titelman continues that “the emotional system includes ‘all the automatic functions that govern the autonomic nervous system’ and can be thought of as ‘synonymous with instinct that governs the life processes in all living things’” (Bowen, 1978, as cited in Titelman, 2014a, p. 26). M. Kerr and Bowen (1988) define the emotional system in this way,

Given the limits of our present knowledge about living systems, it is possible to define the emotional system in only a general way. Defined broadly, the concept postulates the existence of a naturally occurring system in all forms of life that enables an organism to receive information (from within itself and from the environment), to integrate that information, and to respond on the basis of it. The emotional system includes mechanisms such as those involved in finding and obtaining food, reproducing, fleeing enemies, rearing young, and other aspects of social relationships. It includes responses that range from the most automatic instinctual ones to those that contain a mix of automatic and learned elements. Guided by the emotional system, organisms appear to respond sometimes based on self-interest and sometimes based on the interests of the group. (pp. 27–28)

M. Kerr writes that one function of the emotional system concept is to say that all of life is defined by universal life forces. Saying that a human is defined by their emotional system is to say that human behavior is governed more by forces that are common to all of life than are unique to humans. This is a different way of looking at human behavior than beginning with the assumption that human behavior is defined more by what is unique among humans, namely language and other within-species factors pertaining to recorded history and culture described in the first section of this chapter. M. Kerr and Bowen (1988) reflect on the pervasiveness of this sort of automatic response in all of life,

An example of emotionally determined behavior in a lower animal is the activity of a highly stimulated horde of soldier caste ants vigorously responding to intruders into their colony. The ants neither contemplate the meaning of their actions nor harbor strong nationalistic feelings; they simply act. Another example of emotional reactiveness in a lower animal is the teeth baring of a male baboon in response to a stranger. The automatic movement of a plant, a barnacle, or a moth toward a light source is another emotional response. (p. 30)

Bowen (1978) defines it in this way,

Man is conceived as the most complex form of life that evolved from the lower forms and is intimately connected with all living things. . . . Emotional functioning includes the automatic forces that govern protoplasmic life. It includes the force that biology defines as instinct, reproduction, the automatic activity controlled by the autonomic nervous system, subjective emotional and feeling states, and the forces that govern relationship systems. . . . The theory postulates that far more human activity is governed by man’s emotional system than he has been willing to admit, and there is far more similarity than dissimilarity between the “dance of life” in lower forms and the “dance of life” in human forms. (pp. 304–305)

If concepts such as nervous system, vascular system, and others describe physiological structure, then the concept emotional system describes function. It is a concept for organizing the study of behavior from a systems perspective.

The emotional system is a broad concept for natural science research that could bridge the compartmentalization caused by polarities in psychology and psychiatry, such as “psychic versus somatic causes of disease” (M. Kerr & Bowen, 1988, p. 28). “While immunologists, endocrinologists, virologists, geneticists, and other specialists can all describe the activity of pathological processes in the systems they study, they cannot account for that activity adequately” (M. Kerr & Bowen, 1988, p. 29). M. Kerr and Bowen (1988) write,

For example, thinking of the body as an emotional system may enhance our understanding of a clinical problem such as cancer. If the body can be conceptualized accurately as an emotional system, then cancer may reflect some sort of disturbance in the balance of that system. This way of thinking about cancer is quite different from the way of thinking upon which most cancer research has been based. Research on finding the cause of cancer has tended to focus on what is occurring inside the cancer cell. The research question has generally been, “What has gone wrong with this cell to cause it to behave abnormally?” Research based on the assumption that cancer is caused by a defect or disturbance within the cell may eventually provide an adequate explanation. On the other hand, an adequate explanation may possibly depend on being able to conceptualize the body as a biological unit, for example, as a colony of cells. Cancer would reflect a disturbance in the unit as a whole. The disturbance observed within the cell would be a reflection of a disturbance in the larger system of which the cancer-containing organ is a part (p. 29).

Another feature of the emotional system concept is portability from the individual to the relationship system. Papero (2016) describes the emotional system as simultaneously serving two purposes in the individual: the internal regulation of the individual and the regulation of the individual in the context of the relationship system. He claims

Often active below the threshold of a person’s awareness, emotion involves multiple complex interactions of physiology and psychology that deeply influence the individual’s functioning (how the individual responds to the conditions he or she faces). That functioning in turn unfolds in sets of reciprocal interactions with important others, each influencing the other to form repetitive sets or patterns. These patterns can be observed and predicted in conjunction with variables in context. (p. 15)

This two-pronged function of the emotional system in an individual organism is one example of how an individual both defines and is defined by its context. Papero points out the reciprocal nature of emotion as “the force or energy that both produces and results from interaction between discrete living entities and between a living thing and environment” (p. 18). While Darwin defined emotion as the instinctual energy which compelled a single organism to action, Bowen extended this definition to include the interaction of instincts from multiple entities within a single system. Various systems may then interact at different levels, for example, individuals in a family or the various organic systems within an individual or within a single cell. This type of system is “driven” by emotion as defined as the energy for action, is the product of millennia of evolution, and may even adhere to laws of organization fundamental to all of life.

M. Kerr and Bowen (1988) suggest that humans have a tendency to assume that human motives are mostly psychological, that there is a reason for behavior. He suggests that humans assume that an emotional response in an animal, such as rejecting and recoiling away from food, is automatic, while the same response in an adolescent female “is generally ascribed to a psychological conflict” (p. 30). One can ask the human “why” they responded the way they did, but one cannot ask the animal the same question, and so one assumes that the animal has a “how” but not a “why.” M. Kerr and Bowen (1988) write,

Focused as we are on psychological reasons, it is easy to forget that humans, like soldier caste ants and barnacles, are motivated to do many things on the basis of processes that have roots deeper (older in an evolutionary sense) than thinking and feeling. (p. 31)

The emotional system concept underpins Bowen’s definition of the human family as a system. It defines the human family in terms of the integrated, reciprocal relationships between the automatic functioning of the individual and the group. Thus, just as the emotional character of the inputs and outputs of each organ in the body occur through with adjacent organs, Homo sapiens has evolved to function in conjunction with adjacent Homo sapiens as an integrated emotional unit. Through Bowen’s theory describes relationship processes found specifically in the human family relationship system, his broader contribution to biology was the emotional concept itself (Papero, 2015).

As Freud or family therapists applied discrepant concepts to different aspects of human behavior, Bowen sought to develop a theory which could one day integrate all aspects of human behavior into a single frame of reference. The emotional system as an integrative concept was Bowen’s answer to overcoming the problem of using isolated categorical models to describe a single phenomenon. Though the theory itself is not a complete picture of human behavior, he was careful only to add new concepts that was supported by observational facts and possessed a well-defined relationship to every other established concept. Thus, the emotional system concept provides a starting point for organizing research into the relationships between systems within the individual, within the relationship context, and as how these systems relate to systems in other species (Bowen, 1978). Bowen (1978) writes,

Man’s family is a system which I believe follows the laws of natural systems. I believe knowledge about the family system may provide the pathway for getting beyond static concepts and into the functional concepts of systems. I believe that family can provide answers to the medical model dilemma of psychiatry, that family concepts may eventually become the basis for a new and different theory about emotional illness, and that this in turn will make its contribution to medical science and practice. (p. 151)

Differentiation of Self

In biology, the term differentiation has a very specific meaning, which is “the normal process by which a less specialized cell develops or matures to become more distinct in form and function” (Differentiation, 2017). This definition contains a few important implicit points. First, differentiation is a process. Second, it pertains to an individual. Third, it defines something which occurs in that individual in relation to a greater system. Fourth, it indicates that the individual plays a part in an integrated system because “specialized” is a relative term which describes one individual’s function in relation to another individual’s function. Fifth, though it is not directly implied in the definition, it might be inferred that the purpose of the process of differentiation is at least partially guided by the system, and that purpose is to produce a more adaptable system (M. Kerr, 1989).

The pathway of specialization is assumed to be influenced or dependent on the individual’s position relative to other specialized individuals. For example, if a town of 50 has one baker, it is not likely that the next person will become another baker. One might further infer that the type of specialization which occurs is selected naturally for adaptability, an idea consistent with evolutionary theory. Increased specialization requires increased coordination, just as a society with different professional fields requires a common currency to communicate the value of their effort in work. Therefore, differentiation implies adaptability by virtue of increased specialization with increased coordination.

In his early research on schizophrenia, Bowen observed that families grappling with a psychosis were expressing a more intense form of the same emotional process as higher-functioning families (Bowen, 2015). He defined the termdifferentiation of self scale to illustrate the continuum of a family or individual’s dependency on the environment and made it a top-level concept in BT (Bowen, 1978). Papero (2016) writes of the development of the construct,

unlike the psychotic level folie à deux, Bowen observed that this “psychological oneness” can be found not only in severely symptomatic families, but in all families to some degree. And some family members are more caught up in it than others. This observed variation became a part of the foundation for the development of the concept of the scale of differentiation of self, the core of the Bowen theory. (p. 17)

A less differentiated family would require more energy and more resources to survive in the face of pressure from the environment. A more differentiated family would be more efficient in their response to environmental pressures and would have more energy to offer members of the family and also the environment. M. Kerr and Bowen (1988) describe some of the qualities of more or less differentiated families,

Family systems theory also addresses the human’s capacity for cohesiveness, altruism, and cooperativeness. Specifically, the theory attempts to account for the variability in these properties between families. The higher the level of differentiation of people in a family or other social group, the more they can cooperate, look out for one another’s welfare, and stay in adequate contact during stressful as well as calm periods. The lower the level of differentiation, the more likely the family, when stressed, will regress to selfish, aggressive, and avoidance behaviors; cohesiveness, altruism, and cooperativeness will break down. . .The more differentiated a self, the more a person can be an individual while in emotional contact with the group. (p. 93).

This scale was intended in part to show that human families had much in common with other species, from the social structure of ants, to the stress and stampede effect in bovine herds, to mating patterns in primates (M. Kerr & Bowen, 1988; Gilbert, 2006). While the term differentiation pertains to life at all levels, DoS is the process by which an individual human differentiates from their family of origin. It is the core construct of BT.

The scale illustrates the observation that every family functioned along the same fundamental rules, and Bowen defined a scale which had no notion of normal and abnormal, as psychiatry had not properly defined the terms (Bowen, 1978; Nichols, 2017). Poor differentiation is not a pathological pattern developed later in life but is a lack of developmental maturation out of the symbiotic attachment between an individual and the emotional system. Though the effort of differentiation is more intense with caregiver bonds, the process also occurs to a progressively lesser extent between an individual and their siblings, extended family, work, and social systems (Bowen, 1978).

The goal of “coaching” in BT is to assist an individual in expanding awareness beyond their personal frame to develop a level of equanimity with stressors from the rest of the environment (Titelman, 1998/2013). That is, the “work is to differentiate self from one’s emotional systems—the work that makes the difference in lives“ (Gilbert, 2006, p. 29). Bowen (1978) defined two more systems which function along with the emotional system: the feeling system and the thinking system. He writes,

The feeling system is postulated as a link between the emotional and intellectual systems through which certain emotional states are represented in conscious awareness…

The intellectual system is a function of the cerebral cortex which appeared last in man’s evolutionary development and is the main difference between man and the lower forms of life. The cerebral cortex involves the ability to think, reason, and reflect, and enables man to govern his life, in certain areas, according to logic, intellect, and reason. (p. 356)

Bowen wrote that “the terms ‘fusion’ and ‘cutoff’ describe the ways cells agglutinate and the way they separate to start new colonies of cells” (M. Kerr & Bowen, 1988, p. 362). This is one example of the use of terms that port across natural systems, as described previously. The thinking and feeling systems in a less differentiated person are more fused in their functioning. That is, a fused system functions with a higher level of interference from the other. A person who is more fused is less able to access feelings without predominance of thinking and less able to access thinking without a predominance of feelings. DoS is the process through which a person increases their capacity to choose between thinking and feeling as anxiety increases in the environment. Bowen observed that the people who made efforts at differentiating in their own families automatically benefitted from that effort in other areas of life. Therefore, the benchmark of differentiation is seen as how well someone has differentiated from the emotional system in their family of origin (Bowen, 1978). 

There are many descriptions of the various quadrants in the literature (Bowen, 1978; M. Kerr & Bowen, 1988). Some markers of higher differentiation include increased ability to differentiate between opinion and fact, consideration of longer-term consequences of choices, a decrease in the automatic behaviors in relation to the behavior of others, the ability to remain firm in and state one’s deepest convictions without requiring others to change theirs, the ability to retain more access to the thinking system as anxiety increases in the group, the ability to feeling and think something that is not congruent with what another person is feeling and thinking, and many others. These qualities make more differentiated people powerful leaders, even if they are not the ones who are explicitly making decisions for the group (M. Kerr & Bowen, 1988). Alan Gurman, professor of psychiatry at the University of Wisconsin Medical School describes DoS as, “maturation, moral development, the ability to cope with stress, modulate anxiety, and assert yourself without stepping on other people’s toes; in short, being your own person—psychodynamic therapists have been talking about all that for years” (Gurman, n.d., as cited by Wylie, 1991, para. 71). M. Kerr says it is simply, “Differentiation is the ability to think, feel, and act for oneself” (M. Kerr, n.d., as cited by Wylie, 1991, para. 58). “For oneself,” in this sense, points to an individual having more choice over their behavior and less automatic, reflexive, nonthoughtful behavior determined by reaction to the group. M. Kerr and Bowen (1988) write that this individually-determined behavior is not to be confused with individualist and collectivist tendencies in a group, which are factors of human culture. How mature, thoughtful, and intentional a way an individual responds to cultural demands are factors of the emotional system; this is seen as a more basic level of behavior than culture.

Bowen held that the thinking system is ideally informed through, but not ruled by, the feeling system. Differentiation is “not to be confused with avoidance” (M. Kerr & Bowen, 1988, p. 68). An avoidant person is just as reactive to their family emotional system as a person who is fused within it. Therefore, a more differentiated person is able to remain in contact with emotions present in the relationship system while retaining the ability to choose between thinking and feeling, particularly as tension increases. Less differentiated individuals are less efficient coping with stressors. They devote more energy and resources to solve problems that could otherwise be applied to the more productive and enjoyable parts of life (Nichols, 2017; Papero, 2015). Just as less differentiated thinking and feeling systems are less able to function more autonomously, less differentiated individuals are more dependent on their environment for their more basic functions. They are more vulnerable to medical and psychological symptoms but are not necessarily symptomatic so long as the environment is sufficiently supportive (Bowen, 1978).

The original NIMH study (Bowen, 2015), which ended in 1959, provided valuable observations from a quasinaturalistic environment of the inpatient ward. Bowen then moved his research to the Georgetown Medical Center where he continued developing the theoretical system. He published the theory in 1966. By that time, he had accumulated years of experimentation differentiating himself from his own family of origin. When a death occurred in his extended family, he was prepared to seize the opportunity to make an orchestrated move toward differentiation (Bowen, 1978). In a report on this effort, he described his theoretical view of the process and markers of DoS as it applied to the counterbalancing life forces of individuality and togetherness,

Each small step toward the “differentiation” of a self is opposed by emotional forces for “togetherness,” which keeps the emotional system in check. The togetherness forces define the family members as alike in terms of important beliefs, philosophies, life principles, and feelings. The forces constantly emphasize the togetherness by using “we” to define what “we think or feel,” or the forces define the self of another such as, “My wife thinks that . . . ,” or the forces use the indefinite “it” to define common values, as in, “It is wrong” or “It is the thing to do.” The togetherness amalgam is bound together by assigning positive value on thinking about the other before self, living for the other, sacrifice for others, love and devotion and compassion for others, and feeling responsible for the comfort and well being of others. If the other is unhappy or uncomfortable, the togetherness force feels guilty and asks, “What have I done to cause this?” and it blames the other for lack of happiness or failure in self.

The differentiating force places emphasis on “I” in defining the foregoing characteristics. The “I position” defines principle and action in terms of, “This is what I think, or believe” and, “This is what I will do or will not do,” without impinging one’s own values or beliefs on others. It is the “responsible I” which assumes responsibility for one’s own happiness and comfort, and it avoids thinking that tends to blame and hold others responsible for one’s own unhappiness or failures. The “responsible I” avoids the “irresponsible I” which makes demands on others with, “I want, or I deserve, or this is my right, or my privilege.” A reasonably differentiated person is capable of genuine concern for others without expecting something in return, but the togetherness forces treat differentiation as selfish and hostile. (Bowen, 1978, pp. 494–495)

Bowen (1978) gives a clinical example of a husband who stopped giving in to his wife’s demands for togetherness, which impinged on his ability to care for himself. Though the wife protested, the husband was able to hold his ground, and once the wife adjusted to his new position, she thanked him for it. Bowen  considered “this sequence a basic increase in bilateral differentiation which can never return to the former level” (p. 496).

Poorly differentiated people share more self with others, sacrificing more nonnegotiable principles (Bowen, 1988), and so rely more on others to provide them a sense of basic wholeness. A commonly cited example from the NIMH study is where a psychotic patient would belch and the mother would say “excuse me” (Bowen, 1978, p. 6). While this patient and mother represent an emotional fusion, the father was just as distant (Bowen, 1978). Conversely, Siegel (2012) described how infants of depressed mothers participate in the “sharing of such states” (p. 274) and can be as equally unresponsive as their caregivers. In terms of reactivity to stress, Bowen (1988) described differentiation as “the coefficient of personality” (p. 69), in that some personality traits may have a genetic basis, but differentiation partly determines how those traits are expressed. The telltale signs of low differentiation may disappear in situations where one is able to comfortably share self with another. The force of togetherness is dominant for both individuals in these situations, but the comfort is temporary. Eventually one of the two experiences a deficit of individuality, and at that point the closeness becomes too intense to contain within the pair. Under normal conditions the level of intensity is minimal and might represent the small push that compels a person to start an enjoyable conversation with another person. If the tension is intense enough, then one of the two will eventually seek an equally intense togetherness with a third individual in an overly positive or overly negative judgement on the original togetherness, which in turn compels the outside individual to try to get inside one togetherness or another to calm their own anxiety resulting from the outside position. In adults, poorly differentiated people are unable to survive either alone or apart and may cycle between multiple immature relationships in order to feel whole. Bowen termed this cycling between the inside togetherness positions triangling (Bowen, 1978; M. Kerr & Bowen, 1988).


The triangle, not to be confused with the term triad, is the building block in the emotional system. It is a process which occurs between three people, not merely a triadic configuration. Triangles serve an adaptive function in the group as pathways to balance anxiety among the members. Triangles are what makes “strength in numbers” possible and also what makes war possible. In a well differentiated context, an anxious individual can temporarily share their anxiety with others by increasing the level of closeness with one or more members. At a very basic level, this behavior is similar to herding organisms such as cows or arctic muskox huddling together to protect against an approaching predator (Gilbert, 2006). Patterns of triangulation become more fixed in place as the level of anxiety increases beyond a group’s ability to adapt. Chronically anxious groups become more rigid in their functioning and patterns of triangulation eventually become set in stone (Bowen, 1978, 1988; Papero, 2014; Titelman, 1998/2013). If members are fixed on the close side of a triangle they are considered to be fused, and wherever there is a fused dyad there is a third who is isolated (Papero, 2014).

The members on the more comfortable, close side of a triangle or set of interlocking triangles cling to a chosen subjective reality which serves to lessen their own anxiety at the expense of the members on the far side of the triangle or set of interlocking triangles (Papero, 2014). Nichols (2017), a popular mainstream family therapist author adds, “Triangulation lets off steam but freezes conflict in place” (p. 78). These people are in effect ignoring objective facts of their situation in favor of a more subjective, value-oriented representation which supports their need for togetherness as a quick-fix in the moment. An example would be where one person gossips about another using facts which support their own point of view but ignoring other facts which refute it.

There are two ways for an individual to relieve extreme tension in an emotional system, to fuse with it at the expense of self, or to cut off from it completely. Emotional cutoff, the polar opposite of fusion, also relieves tension in the short term but does nothing to change the underlying pattern of emotional reactivity that will affect all relationships in the person’s life (Papero, 2014). Bowen (1988) writes that success in relationship requires harmony between individuality and togetherness, and “harmony [in relationship] requires giving up a bit of ‘self’” (p. 81). He goes on to write that giving up too much self leads to increased anxiety and behaviors such as “overeating, undereating, overachieving, underachieving, excessive alcohol or drug use, and relationships such as affairs [which] are, in part, symptoms of anxiety and attempts to manage it” (p. 87). Ignoring the sensory feedback from these behaviors in favor of maintaining a more subjectively informed sense of self in a fusion can amplify existing proclivities for both medical and psychological symptoms.

Conversely, the dilemma for cutoff individuals is that they will reencounter this pattern in other relationships because they still have very little self to give up in order to maintain harmony in the relationship (Gilbert, 2006). However, while Bowen scholars (i.e., Bowen, 1978; Gilbert, 2006; Papero, 2014; Titelman, 2003) emphasize the benefit of maintaining a connection with as much of the extended family as possible, there is also recognition that emotional cutoff services an important function and may be better than whatever the alternative it protects against. Bowen did promote developing an “‘open relationship to every living relative,’ a goal he believed would do more for enhancing a solid self than anything else he could do in his whole life” (Titelman, 1998, p. 17). The more connections one has in their lives, the larger their support network is (Friesen, 2003; Papero, 2014). In times of stress they can spread their need across more members without relying on a few rigid triangles. These connections operate most efficiently when dyadic relationships are one-to-one, meaning they do not require the involvement of a third person. A network of ideal one-to-one relationships is called an open system (Bowen, 1978).

The Family Projection Process and Multigenerational Transmission Process

Attachment theory is one example of an effort to define a theory of human behavior which moves toward the process of nature as opposed to simply the content of nature. There are many descriptive overlaps with Bowen’s research and attachment theory (Skowron & Dendy, 2004). However, as reviewed earlier, the way that descriptions are conceptualized can significantly impact the courses of action taken in relation to them.

Three main conceptual differences appear to stand out among attachment theory and BT. First, insecure attachment appears to account for low adaptiveness to stress as a deficit in the caregiver’s support of a child while BT accounts for problems as the result of a distribution of anxious focus, generally from higher in the dominance hierarchy to lower in the dominance hierarchy. In BT, this distribution is regulated by the family emotional system and not by any one individual (Bowen, 1978). Second, though a single set of parents are each assumed to have a fixed attachment style, attachment theory does not appear to account for varying health outcomes in siblings which come from the same parents. Third, concepts in attachment theory appear to pertain to the mother–child dyad while concepts in BT pertain to each member in the nuclear and extended families equally and only differ quantitatively (Bowen, 1978).

Bowen observed four mechanisms that a family will use to bind anxiety in order of severity: conflict; distance; overfunctioning–underfunctioning reciprocity; and projection onto a child, which is a particular form of triangling. Conflict requires little explanation but is always explicit. Distance is the opposing reaction to conflict, perhaps described as implicit conflict. It is characterized by decreases in frequency of communication, maintaining physical distance, and others. “Overfunctioning–underfunctioning reciprocity” (M. Kerr & Bowen, 1988, p. 57), sometimes referred to as one-up, one-down, the adaptive position or symptom in spouse, occurs when one spouse begins to invest more resources into the relationship than the other. This can lead to a symptom in one spouse or some other kind of inadequacy in one spouse supporting themselves.

The fourth mechanism, projection onto a child, is the most difficult to resolve. A projection process is a triangle and occurs when anxiety cannot be contained within a single dyad, so the problems in those relationships are projected on a third, usually a child. This particular kind of triangle was important enough to be promoted to a top-level concept in the theory. It is important in the evolutionary component of the multigenerational transmission process which accounts for variation among siblings within the same family (Bowen, 1978).

The parental focus on a perceived or actual problem in a child decreases the anxiety in the original dyad. Multiple triangles can focus on the same person and amplify each other. Bowen observed that a single child could become the object of focus in the parent triangle. This occurs when a problem in a child offers a diversion from anxiety in the parental relationship. The level of anxiety in the parental overfocus is transmitted to the child which in turn increases symptoms in the child. The child, unable to resist the parental overfocus, eventually accepts the part of the parental focus that is out of touch with reality as fact, which obscures the real problem even further. The cycle may be continually intensified as medication and individual psychotherapy are administered to the “sick” patient, or until the individuality force becomes strong enough in the child to function more for themselves (Bowen, 1978).

The concept of a family projection process implies an upper limit to the amount of attention and energy a parent can effectively devote to a child’s needs. Importantly, this notion of an upper limit appears to be absent in attachment theory. The absence of such an upper limit may hypothetically encourage anxious primary caregivers to overfocus on the child in order to maintain harmony in the parental relationship. The child may then develop systems from this overfocus which are only amplified by the assignment of mental health workers to the child’s ensuing symptoms and behavioral issues. Thus, the family projection process expands and includes members of the broader society and fixes the original problem in place. BT predicts that a shift in the balance to decreased child projection and increases in the other three anxiety-binding mechanisms should relieve the symptoms in the child (Bowen, 1978).

The family projection process was visible in Bowen’s original NIMH study. It was observed that when a psychotic child patient would start to improve, the parent would develop a condition. The moment the parent began calling the child “sick,” the patient’s symptoms would reappear. This reciprocity was so predictable that the ward staff would use a change in the patient’s symptoms as a warning for the parent’s symptoms, and vice versa. These observations formed the basis for the concept of the triangle, as it was observed that a change in a symptom was preceded by mother or child alliance with the ward staff against the other (Rakow, 2016).

As previously described, Bowen’s unique methodology used this class of falsifiable predictions which describe how one process would change in relation to a change in another process, but not the content of the processes. The qualitative, process-level prediction was that a change in the parent necessitated a change in the patient, but not precisely what that change would be. These qualitative predictions indicated a jump to studying human behavior using complex systems science before the idea was mainstream. M. Kerr and Bowen (1988) suggest that an amount of objectivity is automatically achieved to get viewing a problem as an unfolding process rather than just the static, categorical concept of that process.

Important observations came from the nurses on the ward, who Bowen trained not to solve the family’s problems but to serve as a resource for the family taking responsibility for working out their own emotional challenges (Rakow, 2016). This was a departure from the typical role of a nurse and was easier for some to adjust to than others. But this “neutral” (p. 154) role allowed the nurses to conduct a more naturalistic observation of the staff. In a review of the nurses’ notes from the NIMH study, Rakow (2016) cites an undated entry logged by Bowen,

Change in the functioning of one family member would be followed immediately by a reciprocal change in the functioning of the family member who was closest attached emotionally, and that this in turn would be followed by reciprocal change in other family members. There was one mother and patient who had no significant emotional ties other than to each other. [The A family] “Each time there was a significant improvement in the patient, the mother would, within a few hours develop a severe physical illness, that could be prolonged and require hospitalization. In another family, the following pattern repeated three times in two months. It involved the mother and patient in the hospital and an adolescent son at home. The patient would get worse, more symptoms of psychosis, the mother immediately become more adequate, decisive, and resourceful, and within the next 24 hours the adolescent son would be picked up by the police for delinquent behavior, like stealing a bicycle, street fighting, and carrying an illegal knife.” [The C family]. (Bowen, n.d., as cited in Rakow, 2016, p. 148)

The selection of a particular child for anxious focus can occur for various reasons, such as overinvestment in the child’s future prior to birth or the development of a symptom in the child. Bowen hypothesized that the child who is caught in a projection process acquires a level of differentiation slightly lower than the parents while the other siblings are relatively free of the emotional oneness and acquire a level of differentiation slightly higher than their parents. Levels of differentiation increase through some lines of inheritance and decrease through others in what Bowen termed the multigenerational transmission process (M. Kerr and Bowen 1988; Gilbert, 2006).

The literature reviewed previously on mainstream developmental theories such as attachment theory reveals no attempt to account for variation outcomes among siblings. As with the other concepts in BT, the multigenerational transmission process is integral to the functioning of families in general, not just abnormal families. Bowen suggested that while some siblings do worse, some do better, and some do about the same as their parents. That is, some lines are decreasing in differentiation, some are increasing in differentiation, and some are maintaining the level of differentiation that they inherited. The variation is accounted for by the degree to which each sibling is freed to differentiate a self within the nuclear family versus remaining mostly dependent on the nuclear family for self. Because a less differentiation is associated with poorer health outcomes as anxiety increases, the multigenerational transmission process may provide a unique conceptualization to account for individual variation in life outcomes in longitudinal research designs (Papero, 2015).

The suggestion that mental illnesses such as schizophrenia or severe intellectual disability could be the inheritance of multiple generations of emotional regression stands at odds with the predominant view that they are caused by physiological factors pertaining solely to one individual within a single generation, or transmitted through DNA. BT does not say that schizophrenia is only a behavioral phenomenon, but offers a roadmap to track the many physiological and psychological dimensions of functioning that influence the expression of that symptomatic pattern in particular.

A concrete, physiological basis for these diseases may indeed one day be found (Nelson et al., 2017). Appropriate therapies may then be developed, but such research will have to sufficiently account for individual variation. It may not be a new hypothetical proposition that emotional symptoms are influenced by the person’s relationship context. However, there appears to be no way to conceptualize and research the complex influence of relationship context on symptomatic expression. As a concept that interlocks with the concepts DoS, triangles, cutoff, and family projection process, the multigenerational transmission process concept provides a way to link broader longitudinal variables from within the relationship context to present day health outcomes, while pointing to concrete pathways to increase family functioning in the future.

Bowen said that through a family diagram of at least three generations, one can very quickly see the transmission of more or less differentiation from parent to child (Papero, 2015). The less differentiated child will find a partner with a roughly equivalent level of differentiation, which is then passed on to their children in a similar fashion. In this way triangles can persist through the generations and beyond memory. Bowen went so far as to hypothesize that a schizophrenic could be produced in as little as three generations (Bowen, 1978). M. Kerr and Bowen (1988) give an approximate timescale for the transmission process to occur,

Although functioning that is stable in most aspects and functioning that is unstable in most aspects are both linked to trends in functioning in a multigenerational family, the rapidity with which changes in levels of functioning (and, consequently, discrepancies in the functioning of family members) occur is variable. Marked discrepancies in functioning can occur in as few as three generations. For example, the functioning of the grandparents of a family member whose functioning is unstable in most aspects may have been fairly stable. Such quantum jumps in functioning are uncommon, however. It is much more common for only mild to moderate discrepancies in functioning to exist after four or five generations. So a fairly stable nuclear family unit can have a descendant who has a chronic schizophrenic level of functioning in just three generations (a quantum jump), but it is more common for such a marked decrease in level of functioning to require five to ten generations to develop. Similarly, a fairly unstable nuclear family unit can, in three or four generations, have a descendant whose functioning is stable in most aspects, but it is much more common for such a pronounced increase to develop over five to ten generations. (p. 223)

Expanded beyond the evolutionary timeline of Homo sapiens, this key concept connects the theory to its roots in evolutionary biology by implying that every individual is the product of their total physiological, psychological, and genetic inheritance. The family diagram, or visible representation of the family tree and the emotional processes through the generations, “reflects the ebb and flow of emotional process through the generations. It defines the vicissitudes of a living organism, the multigenerational family” (p. 306). Thus, consistent with evolutionary theory, the family unit and the species evolve over time to greater or lesser levels of differentiation and adaptation to the environment.

The evolutionary view thus described can appear deterministic at first glance, and the position of the child might appear hopeless as a victim of a pathological process. However, the reciprocal nature of the triangle also implies that each member has the opportunity to move toward differentiation. Emotional reciprocity implies that one person pulling up in functioning automatically impacts the functioning of others for the better. In an often-cited passage, Bowen (1978) sums up the broad, bidirectional impact of a shift in the emotional system:

When any key member of an emotional system can control his own emotional reactiveness and accurately observe the functioning of the system and his part in it, and he can avoid counterattacking when he is provoked, and when he can maintain an active relationship with the other key members without withdrawing or becoming silent, the entire system will change in a series of predictable steps. (p. 436)

This passage points to the predictability of the emotional system at the group level. What is more, a longitudinal view of the family emotional process suggests that any one person’s efforts can have a significant impact on the lives of uncounted multitudes to come.


Bateson, G. (1987). Steps to an ecology of mind. New York, NY: Jason Aronsen.

Bertalanffy, L. (1968/2015). General systems theory: Foundations, developments, applications. New York, NY: George Braziller.

Bowen, M. (1978). Family therapy in clinical practice. New York, NY: Jason Aronson.

Bowen, M. (2015). The origins of family psychotherapy: The NIMH family study project. (J. Butler, Ed.) Lanham, Maryland: Rowman & Littlefield Publishers.

Caskie, P. D. (1994). What kind of system is the family? Family Systems, 1(1), 7-18.

Differentiation. (2017). In the biology online dictionary. Retrieved from

Friesen, P. M. (2003). Emotional cutoff and the brain. In P. Titelman, Emotional Cutoff (pp. 83–107). NY: Haworth Press.

Gilbert, R. M. (2006). The eight concepts of bowen theory: A new way of thinking about the individual and the group.Lake Frederick, Virginia: Leading Systems Press.

Laszlo, E. (1971). Systems philosophy. In G. Midgley (Ed.), Systems thinking (pp. 110–118). London, England: Sage.

Nichols, M. P., (with Davis, S. D.). (2017). Family therapy: Concepts and methods (11th ed.). Boston, MA: Pearson.

Noone, R. J., & Papero, D. V. (2015b). Toward a science of human behavior. In R. J. Noone & D. V. Papero (Eds.), The family emotional system: An integrative concept for theory, science, and practice (pp. 3–14). New York, NY: Lexington Books.

Papero, D. V. (2014). Assisting the two-person system: An approach based on the Bowen theory. Australian and New Zealand Journal of Family Therapy, 35, 386–397.

Papero, D. V. (2015). The family emotional system. In R. J. Noone & D. V. Papero (Eds.), The family emotional system: An integrative concept for theory, science, and practice (pp. 15–28). New York, NY: Lexington Books.

Papero, D. V. (2016). The family emotional system. In R. J. Noone, & D. V. Papero (Eds.), The family emotional system.Lanham, Maryland: Lexington Books.

Rakow, C. M. (2016). Learning from the nurses’ notes for Bowen’s 1954-1959 NIMH project: A window into the development of theory. Family Systems, 11(2), 133-164.

Siegel, D. J. (2012). The developing mind: How relationshhips and the brain shape who we are. New York, NY: The Guilford Press.

Skowron, E. A., & Dendy, A. A. (2004). Differentiation of self and attachment in adulthood: Relational correlates of effortful self control. Contemporary Family Therapy, 26(3).

Stanford. (2011, Feb 1). 24. Schizophrenia. Retrieved from

Titelman, P. (Ed.). (1998). Clinical applications of Bowen family systems theory. Binghampton, NY: The Hayworth Press.

Titelman, P. (Ed.). (2003). Emotional cutoff. New York, NY: Haworth Press.

Titelman, P. (2013). Bowen’s perspective on the therapeutic relationship. In P. Titelman (Ed.), Clinical Applications of Bowen Family Systems Theory (pp. 30–32). New York, NY: Routledge. (Original work published 1998)

Wylie, M. S. (1991, March/April). Family therapy’s neglected prophet: Murray Bowen knew that personal freedom never comes cheaply. Psychology Networker. Retrieved from

Share This

New From Alaska Family Systems

Get The Latest Updates

Subscribe For Notifications on New Posts

No spam, notifications only about new products, updates.

Sign Up For News

Stay current.