The Systems Thinker - Center for Family Consultation's blog

Grandparent Deaths in an Intense Extended Family Symbiosis

Authored by Jim Edd Jones, Ph. D.

In the last two months or so, I have been studying symbiosis; e.g. parent-child symbiosis.  Why?  When I reviewed the quantitative research on families of schizophrenics, it was striking how symbiosis as a factor in development of schizophrenia has been neglected by all the quantitative researchers.  It seemed odd when you consider how strongly Murray Bowen and other schizophrenia researchers of the 1950s had emphasized it. (Hill, Lidz, Mahler, Searles, Wynne, et al.).

While beginning to look at symbiosis, I recalled a 1978 Froma Walsh study about grandparent death that nobody, including Froma, was ever quite able to explain.  Not surprisingly, Bowen theory can explain it.

Grandparent death near grandchild birth

We all have four biological grandparents.  In the general population, about 10% of individuals have had a grandparent die within two years of their own birth. Think about it and you’ll realize that’s a reasonable figure.  I’m one of those 10%.

Walsh, working in a large schizophrenia study at Michael Reese Hospital in Chicago, had noticed in the course of meeting families there that she kept hearing from them more than she would have expected about a grandparent death around the time of the birth of the one who was now in their study as a result of a schizophrenia diagnosis.

She then did this project of carefully gathering information about deaths in the extended families of those in the larger schizophrenia project.  Of the 70 families of schizophrenics for which she got information, 29 of those or 41.4% had a grandparent die within plus or minus 2 years of their birth.  In one comparison group of 25 families from a general nonclinical population, 8% of the offspring individuals had this coincidence of grandparent death near their birth.  In a second comparison group of 45 hospitalized individuals with non-schizophrenic diagnoses, 20% had a grandparent death near their birth.

Walsh talked to another schizophrenia project at Rutgers University and Yale University where Monica McGoldrick was working and asked her to check the records of families in that project.  40.8% had a grandparent death near birth.    Another group (Yates et al. 1989) was studying whether life events of various kinds showed up in certain developmental stages for adolescents who ended up with severe diagnoses.  Without specifically looking for grandparent deaths, it popped up in their data that schizophrenics had more grandparent deaths in the first trimester of the pregnancy that produced them.  This group had no explanation.  They even called the finding serendipitous.

Froma Walsh also never had a satisfactory explanation for her findings, though she had mentioned Bowen’s 1960 “A Family Concept of Schizophrenia” paper.

What’s going on here?  Is it serendipitous to find the same result in 3 different projects?  If not, what’s going on?

Speculation based on Bowen theory

I believe that Bowen theory provides tools for understanding these findings.  For my own clarity, I have sketched in more detail how this might happen, letting multigenerational transmission process, differentiation, death shock wave reaction, and family projection process inform my thinking.

All families have some degree of symbiotic attachment among their members.  The family symbiosis provides some patterned functioning that gives a kind of imperfect stability to the family.  When a child from that system grows up and marries, that will destabilize that family symbiosis to a greater or less degree.  Adjustments have to be made in every case, and the new marriage becomes intertwined with the previous generation’s family symbiosis for the two spouses.  The imperfect stability is restored.

The same process has taken place in the spouse’s family of origin, where the new marriage destabilizes and then restores the temporary stability of the family of origin’s symbiosis.

Likewise, the same process takes place in low differentiation marriages, but can play out differently, due to the low differentiation.  A low differentiation marriage involves two people from low differentiation families of origin.

Consider a hypothetical low differentiation marriage.  Inevitably something happens that disturbs the stability.  Whether a big or small event, low differentiation systems seem primed to have big reactions to events of whatever size.  The intense reactions destabilize the extended family symbioses.  Example events include death of a central figure, geographic move of a member important to the stability of the family symbiosis, a major illness or sharp change in functioning of a person important to the stability, a job loss, war, economic collapse, natural disaster.  Any of these can destabilize one of the two big extended family symbioses of the spouses in our hypothetical marriage.

Any of these events can rupture the extended family symbiosis, resulting in intense waves of reaction throughout the two linked extended family systems.  The two extended family systems frantically try to re-stabilize.  One way is to develop symptoms of all kinds.  Sometimes that’s enough to re-stabilize, sometimes not.  Physical, social, and emotional symptoms may form.  A geographic move of some fragment of the system is another ploy that may work for a while to re-stabilize the system.

The hypothetical marriage we are focusing on is hit by the waves of anxiety from one or the other or both family of origin symbioses.  They scramble.  They distance, or fuse, or get more involved in work, or have affairs, or develop other symptoms.

Or, if they have a newborn, they may project intense helpless neediness on their realistically dependent infant or toddler.  Usually it would be the mother who locks in on the infant with intensity over and above the normal mother-child symbioses from marriages with average level of differentiation.  The father-husband backs off, distances, and supports the intense mother-child symbiosis with a sigh of relief.

If that child has genetic-epigenetic vulnerabilities of any kind (schizophrenia, autism, ADHD, physical deficiency, affective disorder, etc.), the conditions are set for possible eventual (sooner or later) development of severe symptoms in that child.  Associated with that will be a decades long parent-child intense symbiosis that never achieves the usual average kind of resolution of the symbiosis.

This obviously is all very speculative.  But it is one way of accounting for what appears to be a fact, that schizophrenics have had more grandparent deaths that occurred near the time of their birth.  In a low differentiation family system, a grandparent death that occurs near in time to the birth of a grandchild can have a very different impact than what occurs in an extended family system of average level of differentiation.


Bowen, M. (1960). A Family Concept of Schizophrenia. In The Etiology of Schizophrenia, edited by Don D. Jackson., 346-370. New York: Basic Books, Inc. doi:10.1037/10605-012

Walsh, F. W. (1978). Concurrent Grandparent Death and Birth of Schizophrenic Offspring: An Intriguing Finding. Family Process, 17(4), 457-463.

Watt, Norman F., and Armand Nicholi. “Early Death of a Parent as an Etiological Factor in Schizophrenia.” American Journal of Orthopsychiatry 49.3 (1979): 465-73.

Yates, B. T., Fullerton, C. S., Goodrich, W., Heinssen, R. K., Friedman, R. S., Butler, V. L., & Hoover, S. W. (1989). Grandparent Deaths and Severe Maternal Reaction in the Etiology of Adolescent Psychopathology. The Journal of Nervous and Mental Disease, 177(11), 675-680.

Jim Edd Jones


This post was published by

1 Comment on "Grandparent Deaths in an Intense Extended Family Symbiosis"

  • Stephanie Ferrera says

    Dr. Jones brings complex systems thinking to a question that has puzzled researchers for many years. The death of a grandparent within two years of the birth of a grandchild was found to be a much greater than average occurrence in families where the grandchild was later diagnosed with schizophrenia. The death of a grandparent is a discreet, identifiable event, one which would be a factor in disrupting the family stability to greater or lesser degree depending on other variables.

    It is a piece of data that would jump out at researchers. I wonder what other similarly disruptive events could be identified and studied in the previous generations of individuals who develop schizophrenia or other serious illnesses. Dr. Jones give several examples of such destabilizing events. He brings to bear the Bowen theory concepts (multi-generational transmission, emotional shock wave, differentiation of self, and family projection process) that are needed in order to see connections that are otherwise hard to explain.

Leave a Reply

Your email address will not be published. Required fields are marked *