Yesterday, I posted the first part of an article published in the September 2011 issue of Training Journal. I describe how I developed the Onion Model that is at the centre of The Handling Resistance Pocketbook. Here is Part 2 of that article.

Creating the Onion Model - an article by Mike Clayton

From one Model to many

It was around 2004 when a client asked me to extend a presentation skills programme to offer advanced workshops and coaching to small groups. Among their particular challenges, one came up again and again: hostile audiences. The nature of their business and the presentations they needed to give meant that in their audiences were often professionals from other organisations who had a good reason to criticise their work. My client’s team needed to handle adversarial challenges to their work.

In building a toolkit of skills and methods to offer participants, I wanted to give a way that a speaker could diagnose quickly and accurately the nature of the resistance that they were getting. With this, they could adapt their response to handle the challenge effectively. I started playing with the Onion Model to derive a version adapted to resistance to ideas in a presentation. It worked splendidly. Participants were quickly able to discern what lay behind an audience challenge and refer swiftly to effective techniques to handle it positively and courteously. I now had two Onion Models.

The third came much more quickly. It was an ad hoc response to a question about sales objections during a consulting skills course. Asked how to handle objections, I responded that it depends on the nature of the objection. You can guess where I went from there. The group found solutions as I fed their examples into a six-layer framework, and I codified The Sales Resistance Onion Model there and then: they loved it!

The string of onion models is at five now. I pitched The Handling Resistance Pocketbook[5] to the editor of Management Pocketbooks and developed the fourth, for resistance to formal learning, as a part of the outline I prepared. The fifth appeared in last month’s Training Journal, as the Onion Model of Resistance to Engagement[6].

…and back to One Model

The editing process is rigorous for such a tight format as Management Pocketbooks, where every word must count. Sadly for me, the Onion Model of Resistance to Formal Learning was left on the editor’s floor. The biggest challenge, however, was when Ros, my editor, asked for a single, unifying language for all of the Onion Models. The I don’t format meant that each onion had its own labels for each level, reflecting the particulars of the context.

I had never actively thought about the unifying concept behind each level. I knew intuitively that it was there, but by rejecting simplistic labels and jargon, I had avoided taking the time to figure out what they were. But this was a task I relished. It allowed me to wrestle with underlying concepts and play with words: lovely. Some words were easy to find: others were a struggle. I worried for a long time – and still do – that too many people will be unfamiliar with the word enmity (and that I will mis-spell it!) I also worried that people will mis-read content (what’s in it) as content (feeling comfortable). In the text, I could be clear about meanings, but in a simple graphic, there is scope for confusion.


This does allow me to point out that, while the best models need no explanation; very few are that good. Always read the instructions on the tin. A great example is Tuckman’s model of group formation[2]. The words he chose are so simple that people interpret them for themselves, without reading his research.

Is the Onion Model Right?

There are three answers to this. Pick whichever you like.

  1. ‘Yes, it is right’
    It works very well in explaining and predicting resistance. It covers a huge proportion of real-world scenarios. Participants, audiences and clients like its insights and simplicity.
  2. ‘No, it is wrong’
    It over-simplifies a complex set of human responses and misses out many subtle variations. It fails to predict accurately the progress in some real-world situations.
  3. ‘That’s the wrong question’
    Models are neither right nor wrong: they are either useful or not. When we apply a useful model in an appropriate domain, it offers valuable insight and usable predictions. In an inappropriate domain, even the best model will fail.

Model builders try to make the domain in which their model applies, as wide as possible. They try to extend their model to greater extremes of experience. And they also try to pare it down to its simplest form. These are conflicting objectives, so model-building is a balancing task. The extent to which a model is’right’ is the degree to which users find the balance of simplicity and subtlety to be useful.

However, there is one thing I must, as a rational scientist, own up to. The Onion Model is experientially-based, not founded on empirical evidence. It is backed by my experience and that of many people I have observed, spoken with and learned from. But, there is no objective research evidence to evaluate it. I wish there were and would love the chance to work on this, but I have other calls on my time. It�s an awkward confession, but I am also confident in the model’s utility.

The Dangers of Models

Another experientially-based model with little research evidence is Elisabeth Kübler-Ross’s ‘Grief Cycle’[7]. This is coming under scrutiny at the moment and highlights two dangers a model poses. The first applies to experiential models like these and is called the Representativeness Bias. This bias occurs where we find ourselves seduced by a good explanatory story, which fits our understanding of the world, yet is untested by objective observation. Maybe people find that the grief cycle describes their experience because it makes sense of a confusing set of emotions. Maybe that is true of the Onion Model.

The next bias occurs once we have understood a model and are persuaded by it. We can identify experiences that fit into our model and see them as confirming it. The Confirmation Bias can blind us to disconfirming evidence, or at least lead us to interpret it as aberrant, faulty or a special case.

As a rationalist, I need to expose these concerns and invite comment.

Is the Onion Model the Last Word?

Of course not. No model is ever complete and dozens of practitioners will, even now, be trying to tinker with it, overhaul it, or start afresh. There are always exceptions that fall outside the realm of applicability of a model. But William of Ockham would tell us to keep our models simple until we need to make them more complex. Of all the models, hypotheses, theories and laws of nature I studied at university, very few currently have no known exception, nor hints of an exception: perhaps Newton’s laws of motion, or the three laws of thermodynamics. However, as a physicist, there is only one law I would bet on to be truly universal: Sod’s Law.



5. The Handling Resistance Pocketbook, Mike Clayton

6. Resistance to Engagement, Training Journal, August 2011

7. On Grief and Grieving, Elisabeth Kübler-Ross’s and David Kessler, Simon & Schuster (2005)

To get pdf copies of the two Training Journal articles on Vroom’s Expectancy Theory and Maslow’s Hierarchy of Needs, sign up to my newsletter here, and I will email them to you.

Leave a Reply

%d bloggers like this:
Malcare WordPress Security