Ah but such interesting musings!  And mine are much the same... (i.e. musings, and not well enough informed).

Thanks for your time, thought and response.  Stuff to think about is a gift.

Cheers
Gill 

-----Original Message-----
From: Realist and Meta-narrative Evidence Synthesis: Evolving Standards [mailto:[log in to unmask]] On Behalf Of Mickey Sanders
Sent: Sunday, 2 February 2014 8:58 PM
To: [log in to unmask]
Subject: Re: Advice on testing theories - and how programs work.

Hi, Gill.

Well, the idea of distributed agency came to mind when I read these words of yours:

"From this perspective, the mechanism is inherent to the interaction (not the prisoner).  It requires the internal and necessary relations (capacities) of the subject; it requires the whatever-is-provided-by-the-program (the external and contingent component), but the mechanism actually lies in the interaction itself.  Viewed from this perspective, the mechanism did not already exist and lie around waiting to be 'triggered'.  It came into existence in the interaction between the necessary powers of the subject and the necessary powers of the intervention, and in so doing generated its outcome.   One might argue that both the powers and liabilities of the object of interest (the capacities of the prisoner) and the external and contingent relations (program) are context; the mechanism lies in the 'powers and potentialities' of the interaction itself."

You are discussing where powers and potentialities lie, and you are saying that from a certain perspective, it lies in an interaction. That says to me that it these "powers and potentialities" are emergent and distributed, similarly to the way cognition is viewed, in theories of distributed cognition, as being distributed.

Example: When I explain distributed cognition, I usually use the example of doing mental multiplication with Arabic numbers versus with Roman numerals. For most individuals I encounter, they can mentally compute 15*8 while imagining "15*8." The same is not true for most people when dealing with Roman numeral representation of 15 x 8. That representation does not have the same AFFORDANCE ability or power. This, I believe, is what is meant by distributed cognition: Not that the numerals are thinking along with the human, but that the tool (representation in this case) is intimately, inextricably a part of the "thinking system." Of course you can argue about whether the power is solely about the human's ability or whether representations have the power to afford due to contextual factors.

So, Gill, when you mentioned one take that holds that "powers and potentialities" lie in the interaction itself, this sounds like distributed something and emergence, to me. Is it distributed agency? I don't know. But it reminds me of the idea.

One thing burned into my brain is that under RSM, causality is not attributed to programs but to human volitional acts (see http://www.nice.org.uk/niceMedia/docs/Moving_beyond_effectiveness_in_evidence_synthesis2.pdf). I've read Maxwell on the important implications of the difference between how causality is defined/conceived under RSM versus, say RCT.

When I define "mechanism," I have been writing something akin to the following:
"'Mechanism' is defined as subjects' response in light of the resources and opportunities the intervention infuses into the context. It is not solely the subjects' response, nor solely the infusion of resources and opportunities, but the conjunction of the two. Mechanisms do not function deterministically, however . . ."

Now, I don't need to bring in a discussion of distributed agency and emergence to my particular project. But these two concepts do give me a way of thinking about where the potentiality and power lies, in a way similar to how I now think about how cognition works. 

Distributed agency. Affordance. Emergent. Non-deterministically: I can imagine myself drawing a Venn diagram with these 4 items as petals. At their overlap might be C-M, the context-mechanism interaction.

But this is just me, a former math teacher and computer programmer just thinking aloud after having skimmed one article on distributed agency! My writing here might make an expert in these matters cringe! LOL! Just musing aloud.

Please excuse the typos!

Take care!