Exploring Abduction

How often do we really see the generation of new ideas in our organizations or communities?

In a conversation in Agitare Slack recently, Austin Wiggins and I were discussing a recent piece from Dave Snowden called "Abduction through entanglement". The initial building blocks of that piece are two quotes pulled from Nora Bateson 's essay "Ready-ing" starting with Charles Sanders Peirce stating that abduction "is the only logical operation which introduces any new idea; for induction does nothing but determine a value and deduction merely evolves the necessary consequences of a pure hypothesis". This essay will retread much of the same ground that Snowden and Bateson have in the linked essays and others, in an effort to wrap my head around these concepts. If you’re looking for an authoritative view, stop reading my claptrap and go to the sources.?

I’m gonna pause here first to try and unpack that quote from Pierce, because part of what I’m trying to do here is wrap my own head around the difference between induction, deduction, and abduction, so here’s another way of describing the difference between these processes that expands on Pierce’s pithiness:?

This figure is pulled from a 2019 research paper by Nina Baur exploring research methodologies. Here is how Baur describes the difference between these three approaches.?

“Deduction” means that researchers start research by deriving hypotheses concerning the research from the selected theory. Researchers then collect and analyze data, in order to test their hypotheses.

“As illustrated in Figure 1, “induction” starts from the data and then analyses which theory would best fit the data. This simple distinction is another oversimplification in several ways: First, the idea of induction and deductions has been supplemented by the idea of “abduction” (Peirce), which resembles induction in the sense that both start analysis from data and conclude from data to theory (in contrast to deduction). However, induction only draws on existing theories—if no theory is known that fits or can model the data analysis, induction fails. Researchers can only invent a new theory—and this is called abduction” Grounded theory and social-science hermeneutics are the only of the four research traditions which explicitly stress the necessity and importance of abduction, especially as it is the only way of really creating new knowledge.”

Baur then makes it clear that no research process is purely deductive, inductive, or abductive. All researchers in all research traditions must work abductively at some point.

Here’s another view of the difference between these three concepts which can be summarized as follows:?

Given a rule and a cause, we deduce the effect.?

Given a cause and effect, we induce the rule

Given a rule and effect, we abduce the cause

I don’t know how much I prefer this framing as it feels misleading to use the simple term “cause” for something complex and I’m not sure we require the introduction of rules in order to abduce… though perhaps in this context the “rules” integrated into abduction could be understood as how we orient to information through our habitus…???

In Nora Bateson’s essay “Ready-ing”, we’re introduced to another fascinating view of abduction from her father, Gregory Bateson:

“Bateson sees abductive process as the way one context becomes a description of another. Bateson (2002) writes, ‘Every abduction may be seen as a double or multiple description of some object or event or sequence’ (p. 134). The education system can be seen as a description of the economy, the health system, the law, the history and the family culture. The same sort of transcontextual description exists in the meadow or the forest through the ecological interdependencies. The moss is a description of the tree, the bacteria in the soil are describing the forest, and the insects that live in the forest are describing the other flora and fauna.”

I recently got the opportunity to experience a small taste of what Nora Bateson calls “Warm Data” in a virtual workshop–the online version of her “Warm Data Labs” that she calls “People Need People”. In her opening remarks for the workshop, she made a distinction between two types of data:

• Warm Data: trans-contextual, not yet dissected and disconnected by invented distinctions of category and scientific discipline, still entangled with countless other parts and pieces of perception, experience, and actuality. Warm data is "information that is alive", the kind we share with one another in stories, testaments, and conversations without removing it from context
• Cold Data: Lifeless, having been removed from context, dissected with abstractions and classifications. Most of the data we work with is cold data.

A conversation that weaves in and out of the lived experience, perceptions, and memories of the participants discussing a handful of things creates data that is warm and alive. It is like a tangled knot of information that can’t be cleanly separated from who we are, what we experience, and how we perceive the world. It is an ecology of information–alive and complex rather than cleanly severed from context and sorted into categories. I’m reminded of G. Bateson’s “The moss is a description of the tree”. When I’m having a rich discussion with someone about a handful of topics, the conversation and understanding is embodied and visceral. Where there is a difference of experience or perception, we use metaphor and exploration of the conceptual flora surrounding a thing to try and articulate its description. The shape or description of concepts is revealed by the concepts and information that grow on and around it. The information we generate together isn’t clean or categorical. It is contextually-bound, and only really fully knowable by those who generated it. To really share that information in its warmest form would require us to generate something new with the people we are inviting in.

The concept of "new ideas" popped into my head the other day as I was walking my dog and thinking about how often we actually create new information in institutional contexts. Most of what we do when we're introducing an idea that feels "new" is simply bring generalizations that we've found (cold data) to a new context and introduce them to the static assumptions (cold data) that form the social structures that we occupy.

For example, if someone reads Dan Pink’s popular book Drive and his framing of motivation makes enormous sense to them, and they bring it back to their organization as a lens to start employing in how they should lead people, this might feel like new information is being brought in, but a generalization or abstraction can't really be new information can it? It's simply a scaffolding upon which to drape the current data, most of which is cold. I wondered if it could be argued that this is what Bateson refers to as one context becoming a description of another, but then… what’s happened isn't a description of an object, event, or sequence… it’s a description of an existing abstraction. It’s the layering of one generalization on top of another, and the result might feel revelatory, but it can’t be abductive without direct contact with the target of those generalizations, and if you’re starting with generalizations, isn’t this just deduction or induction after all? We’ve decided we like the descriptive power of Pink’s “Autonomy, Mastery, and Purpose” framing of motivation and are therefore going to drape it over the context within which we engage, not because the context generated this concept, which would be abductive, but because it feels like a useful way of resolving or engaging with the context. An abstraction-first approach can’t be abductive, and one reason it’s less powerful than an abductive approach is precisely because it limits, categorizes, and filters information in a way that can only validate/invalidate initially-held assumptions (deductive) or determine values within a pre-defined set of classifications… say by determining the degree of autonomy, mastery, and purpose of the individuals within a unit (inductive).?

I think I’m getting this right, but honestly I can’t say I’m 100% confident, because I have always found these terms extremely confusing. So part of the reason I’m writing this is to serve as bait for thinkers I respect to come along and correct me (hopefully gently).

Perhaps this all seems like an exercise in pointless academic navel-gazing to you, but I promise it’s all leading somewhere that I’m convinced we should all find useful.?

So if we start with abstractions, we’re not only introducing nothing new, but potentially introducing an obstruction to seeing real information within our domains of practice. To actually get new information in its warm, trans-contextual, unfiltered form, we need to engage with it directly, not through the obscuring lens of some abstraction. I’ll be honest, the prospect feels a bit impossible given that many abstractions might be baked into our consciousness in the form of habitus. Perhaps our ability to access warm data increases when we engage in abductive practice with others whose habitus might deviate from our own… This speaks to the importance of diversity…

I realized as I was writing this that the idea of “new information” was on my mind from a recent watching of the film “Good Will Hunting”. There are two scenes in the movie in which abstractions are pitted against the “truth” of embodied and trans-contextual knowledge. The first is the scene where Matt Damon’s character (Will) schools a Harvard student in the bar by first revealing that all of his stated “opinions” are simply the recycled opinions of other thinkers, making none of his “intellect” the product of actual original thought but merely rote repetition. Will makes it clear that the degree that the Harvard student is earning could be acquired for $1.50 at the library. The college kid’s retort that he would always occupy a higher socio-economic class than Will (because in the end it’s about the degree, not the knowledge) is met with “Yeah but at least I won’t be unoriginal”. The message from this scene is clear to me–repeating others’ ideas isn’t the same as thinking, even though it’s what our culture tends to reward.

The second scene on this theme sees Robin Williams’ character putting Will in his place by (ironically) describing how no amount of book-reading will ever give a person the knowledge gained through experience.?

“Michelangelo? You know a lot about him. Life’s work, political aspirations. Him and the pope. Sexual orientation. The whole works, right? I bet you can’t tell me what it smells like in the Sistine Chapel. You never actually stood there and looked up at that beautiful ceiling. Seeing that. If I ask you about women, you’ll probably give me a syllabus of your personal favorites. You may have even been laid a few times. But you can’t tell me what it feels like to wake up next to a woman and feel truly happy. You’re a tough kid. I ask you about war, you’d probably throw Shakespeare at me, right? ‘Once more into the breach, dear friends.’ But you’ve never been near one. You’ve never held your best friend’s head in your lap and watch him gasp his last breath, lookin’ to you for help. If I asked you about love, you’d probably quote me a sonnet, but you’ve never looked at a woman and been totally vulnerable. Known someone that could level you with her eyes. Feelin’ like God put an angel on Earth just for you, who could rescue you from the depths of hell. And you wouldn’t know what it’s like to be her angel, to have that love for her be there forever. Through anything. Through cancer. And you wouldn’t know about sleepin’ sittin’ up in a hospital room for two months, holding her hand, because the doctors could see in your eyes — that the terms ‘visiting hours’ don’t apply to you. You don’t know about real loss, ’cause that only occurs when you love something more than you love yourself. I doubt you’ve ever dared to love anybody that much.”

These two scenes were on my mind as I thought about how much we actually generate new information in an institutional context. How much of this information contains experience? How much of it is simply the rote recitation of abstractions borrowed from other thinkers? I should be clear that I consider little of the information I'm presenting here as new information, since most of it is simply relaying the thinking of others... except the parts where I connected things that weren't initially connected, that connected through my embodied experience... walking the dog, thinking about "Good Will Hunting", remembering the "People need people" workshop, thinking about ecologies everywhere as I'm currently reading Robin Wall Kimmerer's book Braiding Sweetgrass, talking with Austin about essays we're sharing with one-another... stewing in a tangled web that forms an ecology of ideas--the ways that these things become descriptors of other things... according to G. Bateson that is arguably abductive and therefore presents something new...

So how do we get something new? How do we generate new information? I think it should be obvious to us just how often we allow ourselves to engage in abductive processes–those that don’t put the question “What matters most” before the question “What’s happening?” (or in Nora Bateson’s workshops: “What’s Continuing?”). Liberating Structures offers us the framing “What, So What, Now What”, and I think it's really useful to ask how often we really start with "what?"

To be even more specific… think about the last time an actual open, exploratory, psychologically-safe discussion occurred in your group with a diversity of experience and thought represented (if ever). Did the discussion begin with statements of important abstractions from a leader? If so, then you already limited the amount of warmth and life that that data could hold. If it was overly contextualized, it couldn’t possibly be trans-contextual because people kept pulling it back to the original framing. If it was in service of some kind of specific, pre-determined output, then it couldn’t really be “warm data” that you created because beginning at the end started you off with assumptions about the “now what” and that’s not the order of abduction.

Of course things are trickier than I’m letting on here. These are imperfect models I’m putting in front of you. I’m just thinking, and I hope I’m helping you think as well. How often do you generate new information in your organization? Is it already classified/categorized by the time you see it? Do you only see it because of categorizations or abstractions you’ve settled on? If so… just how new could it possibly be??

P.S. The idea of abduction pairs nicely with exploring information & experience without attempting to make sense of it, something I wrote about a few months back.

P.P.S (from the comments) One of the really interesting things about the Warm Data Labs experience was how the practice and process itself didn't include a "so what" and "now what" element. It was specifically about creating space and stimulus for abductive exploration/warm-data weaving to occur and then the "what happens next" was really left to the individuals/groups to determine for themselves. Clearly, existing abstractions can't be fully removed from this process (given that they occupy us as part of our habitus) and they will certainly come into play afterwards, but as Baur notes, no research process is purely deductive, inductive, or abductive. I think one of the most powerful things we can do is intentionally include abduction in our sense-making, and allow that practice, which includes diverse lived experiences and diverse viewpoints to have influence over the abstractions we choose to go forward with, and those we choose to leave behind.