Updated:
If you have worked with Bowen’s family systems theory, you may have noticed that some of its core concepts pull in different directions. These tensions are not errors — they reflect genuine complexity in the theory. Active Inference, a framework from contemporary neuroscience, offers a way to revisit these tensions through one of its foundational concepts — the triangle.
Tensions in Bowen’s Triangle
- Bowen’s theory presents differentiation of self and chronic anxiety as two distinct variables — the main variables of the theory — yet treats them as correlated: as one increases, the other decreases. Are there one or two variables in the triangle dynamics?
- Bowen’s theory holds that the more individuality force, the better — yet it also presents the optimal state as a balance between the life forces of individuality and togetherness ”neither force overriding the other” (Bowen 1978, p.276). These arguments are valid in the contexts where they are used, but seem logically contradicting each other. How do you reconcile them?
- The level of differentiation is, according to Bowen theory, largely stable across an adult lifetime and emotional functioning is essentially instinctual — yet the entire clinical approach assumes that people can consciously develop their differentiation. How can one actually work on differentiation given those constraints?
The Reconstruction in Brief
The following readings address the three tensions outlined above and are expressed, when possible, in the language of Bowen theory. Their meaning, however, reflects a formal reconstruction of the triangle — one that draws on a contemporary Active Inference framework to clarify the logical structure of Bowen’s original ideas. A logically consistent theory can, in principle, be given a formal expression. This, however, does not call for measurement or calculation in practice — a position Bowen himself held (1988, p. 387, see also Kerr 2020, p. xix).
This interpretation of the triangle emphasizes two structural features of human functioning: the temporal distinction between long-term development and short-term adaptation, and the predictive — rather than reactive — nature of behavior.
- Differentiation of self and chronic anxiety are the two main variables of Bowen theory — and they are formal complements: as one increases, the other decreases. This complementarity reflects the fact that they are not truly independent constructs, but two expressions of a single underlying variable — in the reconstruction, the minimization of free energy, which is the same process as the maximization of self-evidencing. Their difference lies in location. In a triangle, each person’s internal variable drives their behavior, and that behavior is an external variable for the other two. For one individual, the variable appears as differentiation of self, which refers to how accurately they meet their environment, both in observing it and in acting within it, and as its complement, internal chronic anxiety. The behavior of the others, driven by their own level of differentiation, reaches the individual as external chronic anxiety in the family system. Seen this way, the apparent contradiction dissolves: Bowen’s theory does not present two separate variables, but one variable expressed as two complementary aspects of a single process.
- The life forces are often understood as human needs, and Bowen’s own language sometimes invites this reading. Here they are understood differently, following his description of them as ”inherited from human’s phylogenetic past, that he shares with the lower forms, and that governs the subhuman part of man” (1978, p. 422). In Bowen’s account, the life forces operate in the emotional system and drive its processes of adaptation in the relationship system, whether or not the person is aware of them or wants them. Togetherness is here understood as openness to the influence of others, and individuality as holding on to one’s own beliefs. The apparent contradiction between ”more individuality is better” and ”balance is optimal” dissolves once we recognize that the two claims refer to different time scales. Over the long term, the life forces point toward a more developed self: as the person’s understanding of life grows, the beliefs held become better grounded, and there is less need to follow others. In the moment, functioning is best when togetherness and individuality are in dynamic balance, neither overriding the other. Bowen (1978, p. 276) illustrates this with a 50-50 balance, but the point is not rigid equality. The balance is not an even split across all beliefs: the more a belief – whether true or not – is tied to survival, the more firmly it is held. Too much of either force, driven by chronic anxiety, leads to over- or underadaptation. Bowen’s arguments are not contradictory — they are valid in their respective contexts. The confusion arises when a statement about the long-term trajectory is read as a statement about the momentary state or the other way around.
Kerr (1988, p. 95) has noted that differentiation and individuality are often used interchangeably, although differentiation refers to a process and individuality to a life force. Kerr (2019, p. 72) later describes the two as attempts to conceptualize the same idea in different contexts: differentiation within the individual, and individuality within the relationship system. A good basic level of differentiation allows a person to act on either individuality or togetherness as the situation requires (Kerr 2019, p. 68). - In principle, between stimulus and response there is a moment where the individual can either follow the automatic emotional response or engage a more deliberate one. Yet this formulation conceals a tension: if emotional functioning is instinctual and non-conscious, that moment is by definition not under the individual’s control, as Bowen theory implies.
A relationship system can also be calm without the individuals in it being accurate. When the pattern is stable, nobody is surprised, and nothing invites anyone to question it. Yet those involved may still be partly out of tune with their own bodies and with the rest of nature, in biological, psychological and cultural terms.
In the reconstruction, the individual does not control the moment between stimulus and response. The moment arises when the behavior of others disturbs the stable pattern and makes the automatic response less certain. The relational system can either strengthen or weaken the automatic response. Whether a more deliberate response can then take over depends on the individual’s capacity to tolerate uncertainty, and this capacity depends on how well the individual can tell which matters are crucial for survival and which are trivial. The more finely the individual distinguishes between situations, the easier it is to tolerate uncertainty in trivial matters and still act decisively in the important ones. Both aspects of the main variable are necessary: the relational system creates the conditions (external aspect), and differentiation of self (internal aspect) determines what becomes of them.
This also shows where differentiation can be worked on. It is in these moments, through observing the situation more accurately (insight) and acting on that observation (behavior), that the level of differentiation can slowly change, within the range set by the family of origin.
These readings are brief by design. The reasoning behind each of them is opened up step by step in the newsletter, published roughly every two months. It covers whether the triangle holds one variable or two, why balance rather than simply more individuality is optimal in the moment, and how differentiation can be worked on within the constraints Bowen theory sets. You are welcome to subscribe at the end of this page.
References
Bowen, M. 1978. Family Therapy in Clinical Practice. Jason Aronson.
Bowen, M. 1988. Epilogue: An Odyssey Toward Science. In M. E. Kerr & M. Bowen, Family Evaluation: An approach based on Bowen theory. (pp. 339–386). W. W. Norton.
Kerr, M. E. 1988. Differentiation of self. In M. E. Kerr & M. Bowen, Family evaluation: An approach based on Bowen theory (pp. 89–111). W. W. Norton.
Kerr, M. E. 2019. Bowen theory’s secrets: Revealing the hidden life of families. W. W. Norton.
Kerr, M. E. 2020. Foreword. In Keller, M. N., & Noone, R. J. (Eds.). Handbook of Bowen family systems theory and research methods: A systems model for family research. Routledge.
Explore the concepts interactively
To demonstrate the reconstructed concepts of the triangle, the following interactive tool visualizes the relationships between differentiation and the vertical and horizontal dimensions of the life forces — for one agent only, not the triangle as a whole. Feel free to play with the variables. The tool opens in a new window; use the instructions below to read the visualization.
Instructions for the interactive tool
- The vertical axis represents differentiation of self as an internal variable. The higher the position, the more accurately the agent meets their environment, both in observing it (objectively, in Bowen’s terms) and in acting within it. The range of differentiation is set by multigenerational transmission using the first slider, and the red tick marks on the axis show this range. The second slider moves the gray dot, which shows the current level of differentiation within that range.
- The horizontal axis represents the balance of the life forces in the relationship system in momentary adaptation to others. The center is the optimal state. The labels below the axis show how sustainable the adaptation is: optimal at the center, sufficient halfway toward each end, and unsustainable at the ends. Movement toward either end indicates a bias driven by chronic anxiety — toward excessive togetherness on the left (over-adaptation), or excessive individuality on the right (under-adaptation). Deviation from the optimal state corresponds to the concept of pseudo-self, and the bias runs in either the functional self direction (excessive individuality) or the functional no-self direction (excessive togetherness).
- The dashed diagonal line shows self-evidencing over time: as differentiation increases, the individuality force — understood as trusting one’s internal model of beliefs — grows toward its theoretical optimum, reflecting a long-term or developmental view of individuality. How far this development can go depends on the range set by the multigenerational history of the family. The dark blue dot marks where the agent is on this line at the current level of differentiation. Its horizontal position does not show the momentary balance; that is shown by the turquoise markers.
- The dashed lines from the gray dot to the turquoise tick marks on the horizontal axis show how the level of differentiation sets the deviation zone: the lower the differentiation, the wider the zone. The third slider sets the chronic anxiety of the system. The turquoise dots show how far this anxiety pushes the agent toward over- or under-adaptation within the zone, and the red line between them shows the degree of external chronic anxiety — the pressure coming from the chronic anxiety-driven behavior of others in the relationship system. The red segment on the vertical axis, above the gray dot, shows the degree of internal chronic anxiety — the complement of differentiation within the agent.
Agency Triad
Triangle is the name Murray Bowen gave to a foundational concept in his family systems theory. The Agency Triad is a reconstruction of that concept — not a replacement, but a formal interpretation of what the triangle describes and how its core concepts relate to each other.
The term ”Agency Triad” is deliberate. Active Inference, the contemporary framework informing this reconstruction of Bowen’s triangle, is precisely a theory of agency. Kerr (2019, p. 268) also refers to having a self as having agency — a connection that points toward the central process of the Agency Triad. From the perspective of research on professional agency, dimensions of agency align naturally with both Bowen theory and Active Inference.
The ”Agency Triad” acknowledges the distance from the traditional interpretation of Bowen’s triangle while recognizing the debt to Bowen’s foundational work — this reconstruction is a reinterpretation, not a revision.
About Petri Ruotsalainen
My encounter with Bowen’s triangle goes back to the late 1990s, when I studied theology at Baylor University in Texas and read Generation to Generation by Edwin H. Friedman. The triangle later became the foundation of my work as a trainer and consultant when I started my own practice in 2001, focusing on leadership and collaboration. After all these years, Bowen’s triangle still manages to surprise me. It is relevant, has strong explanatory power, and is a profound model of behavior.
In 2022–2023 I completed the Postgraduate Program in Bowen Family Systems Theory and Its Applications at the Bowen Center. I am currently completing a doctoral dissertation at the University of Jyväskylä, Faculty of Education and Psychology, where I study professional agency — particularly its embodied and emotional dimensions — alongside my practice. A central part of my research is theoretical: in collaboration with the Faculty of Information Technology, I have been developing a mathematical formalization of Bowen’s triangle.
Knowing and Becoming
Bowen’s theory is sometimes read as a warning against shared thinking — as if dialogue itself were a form of fusion. I take a different view. Humans survive as a group; learning and development happen naturally in collaboration. Working toward differentiation does not require avoiding reasoned argumentation or shared reflection. On the contrary: dialogue — the exchange of thinking based on one’s observations of relationships — is a laboratory for noticing fusion and practicing differentiation, much as engagement with one’s family of origin is the natural ground for strengthening differentiation.
Theory as propositional knowledge needs to become embodied — lived rather than merely known. As the Kantian dictum is often paraphrased: theory without empirical data is empty; empirical data without theory is blind. Yet theory sharpens observation, and dialogue is a way to deepen its understanding.
My practice focuses on leadership and collaboration development within the framework of professional agency — the agency triad being a central tool. If you are looking for professional support, I offer online coaching and consultation.
You are more than welcome to contact me whether you want to share your thinking about the Bowen’s triangle and its reconstruction or discuss how it might serve your professional development. My email address is petri@nonlinear.fi.
Would you like to read more?
My newsletter features reflections on the Agency Triad, published roughly every two months. You are welcome to subscribe. Here you can see an example of the topics covered in the newsletter.
Please check your spam folder if you do not receive a confirmation email within a few minutes. You need to confirm your subscription to start receiving the newsletter. Once your subscription is confirmed, newsletter editions can only be accessed via the link in the newsletter email.