This was an unusual commission for us, in that it’s outside of the commercial sphere, but it provides a good illustration of our systems-synthesis process, together with an application of part of our energy-dynamics model to a rather extreme context.
The client was a new online academic journal, Nuance International Journal of Family Policy and Related Issues. Two of the journal’s editorial team had seen some of my previous work on power interactions in interpersonal relationships (see the book-chapters ‘Power and Fear‘ and ‘Use and Abuse‘ on the Tom Graves section of this web-site), and asked me to review, critique and, where appropriate, suggest revisions for the commonly-used ‘Duluth’ methodology for resolution of domestic violence. It was not something I could do solely on my own; so in typical Xio style, I built up a network of colleagues who were willing to help, including a retired Family Court lawyer and several academics.
In accordance with the systems-synthesis process, we went back to first principles:
- what exactly was the Duluth methodology supposed to do?
- what was the purpose?
In other words, what were the nominal-requirements for the Duluth methodology, viewed as a total-system? (In this case, it was a total-system with a zero, or almost-zero, information-technology component, but the principles remain the same.) As usual, the fundamental base-requirement was very simple: in this case, “assist perpetrators of domestic violence in changing their behaviour, to prevent or minimise future recurrence of such violence”.
But as we looked in more depth, it became clear that apparently ‘obvious’ distinctions such as those between ‘perpetrators’ and ‘victims’ are often subjective and arbitrary, and usually misleading (in this case, because life-histories show that most people, in most circumstances, are actually both); and crucial terms such as ‘power’, ‘responsibility’, ‘abuse’ and ‘violence’ are often used in ways which are ambiguous, incomplete and self-contradictory. By following the systems-synthesis methodology, and its insistence on ‘clean’ criteria (in other words, criteria without arbitrary boundaries), we arrived at the following assessment-criteria:
- 1. The model must be underpinned by theory based on proven objective fact
- 2. The model must provide clear, consistent, experiential definitions of concepts such as power, responsibility, abuse and violence
- 3. The model must address abuse and violence between any parties
- 4. The model must address all abuse and violence between involved parties
- 5. The model must address all forms of abuse and violence
- 6. The model must provide active support and incentives for behaviour change
- 7. The model must address issues in a manner which is consistent and fair to all parties
We then assessed the published Duluth methodology against these criteria. (The Duluth model is formally described in Pence and Paymar’s book Education Groups For Men Who Batter: The Duluth Model, Springer Publishing, New York: 1993.)
We soon found that the original Duluth model has serious failings in every case: although in principle the model’s ‘re-education’ concept ought to work well within the context, the problems with the design and implementation are so severe that any successes are more likely to occur in spite of the model than because of it. For example:
- for more than a decade, almost all of the supposed ‘facts’ on which the original model is based have been shown to be mispresented, misleading or just plain wrong;
- it only addresses violence by males, in fact explicitly refuses to address any violence by females (hence, for example, making the model unusable for resolving lesbian violence, which statistically is the most common form of inter-partner violence);
- it addresses violence issues in a manner which is selective and methodologically indefensible, and fails to address at all some serious issues such as sexual abuse;
- the model essentially describes violence as ‘power’, making it almost impossible to create constructive change;
- its only method for creating change is blame and disempowerment;
- it has no real criteria for assessment of ‘success’; and
- its procedures and assumptions are clearly inconsistent, and clearly unfair not just to men, but to women too.
In short, not so much a flawed methodology as a potentially-dangerous, almost unusable mess.
In a commercial environment, obviously, something as badly designed as this would cripple a company very quickly indeed. But we were told that, amazingly, the present Duluth methodology is actually mandated by law in more than half of the US states, and in several Australian states as well, as the only legally permitted method for supposed ‘resolution’ of domestic violence. What’s really worrying is that no-one involved in those states appears to have thought to carry out what seems, to us, to be an essential and obvious process of review…
The next step in systems-synthesis is to compare the nominal-requirements with the effective-requirements which the system actually implements – in other words, reverse-engineer the effective requirements from the details of the actual implementation. This exposes covert-requirements which, if they conflict with the nominal-requirements, need to be addressed – otherwise their effects are likely to reappear in any revised system.
Covert-requirements usually arise from subjective assumptions, circular reasoning and similar error-prone habits of thought: and it was clear that the Duluth model was littered with them, almost everywhere in its design. Some were obvious, such as its arbitrary assumption that violence was an exclusive ‘male’ characteristic, and the circular reasoning used to promote that assumption; some were more subtle, such as the assumption that systematic disempowerment of supposed ‘perpetrators’ automatically empowers supposed ‘victims’. Each of these conflicting covert-requirements reduces the system’s scope relative to the base-requirement, in this case to a point where the effective performance of the system was almost nil. (Against a more cynical base-requirement, such as “the system shall ensure that, wherever practicable, males shall be punished for being male”, the system’s performance could be said to be very high… although, as in most cases where covert-requirements dominate a system’s implementation, it is doubtful whether Duluth’s designers consciously intended such a result!)
Systems-synthesis uses a wide range of techniques to identify and reduce the impact of covert-requirements. Some of these, such as the reductio ad absurdum, are traditional and well-known; others are less well-known but logically obvious, such as the use of language-deconstruction techniques against deconstruction itself.
The formal paper published in Nuance Journal showed the end-result: a revised model and methodology for violence-resolution which can be used in any context, from intra-personal (self-abuse), through one-on-one interpersonal (as seen in domestic violence), to transpersonal (between groups, and even between nations). That rework of the Duluth model (see ‘Duluth redesign‘ in the Tom Graves section of this website) now forms a core part of Xio’s energy-dynamics model – not least because exactly the same issues can be seen, in a lesser or more subtle form, in so much of ‘office politics’.
With the paper complete, we returned the issue back to the academics. That debate is well outside of Xio’s scope – and with so much blame and anger dominating so much of gender-politics, we’re keen to keep it that way! The original Duluth model was so poorly designed and implemented relative to its nominal base-requirement that it’s possible that almost anything would have been an improvement; but we’re pleased to have been told by some of the practitioners working in the field that the revised model we’d presented in Nuance Journal is turning out to be one of the few tools which actually helps – rather than hinders – resolutions for that particularly bleak social problem.