🔮 Uncertainty as Doubt (Part 2)

Good morning!

At the Uncertainty Project, we explore models and techniques for managing uncertainty, decision making, and strategy. Every week we package up our learnings and share them with the 2,000+ leaders like you that read this newsletter!

Last week, we started a tour of how uncertainty has been discussed in economic circles over the years, and how that has impacted ideas for strategic decision making in business. This week, we finish this tour of ideas from Amar Bhide’s “Uncertainty and Enterprise”

Thinking about uncertainty as a mental state of doubt, as Amar Bhide suggests, opens up new tactics for decision making.

What does it take to reduce our doubts, in the name of reducing uncertainty? We’ve explored several angles:

• Agreement - involving other people (covered last week)

• Justification - making a case (covered last week)

• Evidence - volume, quality, and interpretability (covered last week)

• Ambiguity - degree of interpretability (covered last week)

• Large Organizations - and their routines (covered this week)

• Narratives - two modes, and good form (covered this week)

In general, Bhide says we seek to reduce our doubts by gaining agreement with others, using narratives that provide justifications based on evidence - that is often steeped in ambiguity. In large organizations, this is often supported by specific routines.

Let’s pick up where we left off.

On Large Organizations

Organizations of different sizes can (and do) take different approaches to uncertainty. Bhide frames enterprises by the magnitude of the investments they make: 

He notes that large organizations are naturally more risk-averse because:

“Large companies that undertake complex, long-gestation megaprojects also require evidence supporting the prospects for commensurately large payoffs. In my terminology, the evidentiary requirements limit tolerances for irreducible market uncertainty.”

“Large companies routinely reject uncertain ideas that VCs, angels, or informally financed entrepreneurs are willing to try out.”

But why are large organizations predisposed to avoid this?

Bhide taps into business history, surveying some landmark studies of large organizations (in the US), to show how they shaped many of the norms and conventions we still see today:

In Alfred Chandler’s “Strategy and Structure” (1962), he explained how “strategy dictated organizational structure.” He said this design has two aspects:

1. “Lines of authority and communication between the different administrative offices and officers, and

2. Information and data that flow through these lines of communication and authority.”

He added, “Such lines and such data are essential to assure the effective coordination, appraisal, and planning so necessary in carrying out the basic goals and policies and in knitting together the total resources of the enterprise.”

This sounds obvious by today’s standards, but it was new in the post-war period.

Alfred Sloan, the chief executive at General Motors (1937-1956), was a leadership pioneer who developed a tradition of “selling ideas, rather than simply giving orders.” All levels of management had to “make a good case” for their proposals; the managers who wanted to “operate on a hunch” would “find it hard to sell his ideas to others.” He felt that it protected the company against “ill-considered decisions, by assuring that basic decisions were made only after thorough consideration by all parties concerned.”

This emphasis on group decision making was noticed by John Kenneth Galbraith when he wrote “The New Industrial State” (1967).

“Decision in the modern business enterprise is the product not of individuals but of groups. The groups are numerous, as often informal as formal, and subject to constant change in composition. Each contains the men possessed of the information, or with access to the information, that bears on the particular decision, together with those whose skill consists in extracting and  testing this information and obtaining a conclusion. This is how men act successfully on matters where no single one, however exalted or intelligent, has more than a fraction of the necessary knowledge.”

Galbraith also made a great distinction between decision making for “running the business” and decision making for “changing the business”:

“Just as large corporations have low but not zero tolerance for missing information, their procedures for evaluating new initiatives are not algorithmic or formulaic. The distinction between procedures for ‘operating’ and ‘enterprising’ choices is notable. Operating rules - using computer-generated credit scores to issue credit cards, for example - can be rigidly formulaic. Enterprising routines, which include interpreting ambiguous contextual data, aim for ‘procedural rationality’, not substantive invariance.”

But these “enterprising routines”... they can get heavy. What’s behind them? Bhide argues that  - whatever it is - it’s not the result of mainstream economic theory:

“Why do large companies establish such elaborate and collectivized routines rather than use the more streamlined procedures of VCs and nimble pre-IPO businesses? Why did companies like General Motors, with reputations for obedient deference to organizational rank, also develop traditions for consensual decision making? And why do high-tech companies that claim to detest bureaucracy adopt these collectivized practices as they mature? These questions do not arise in economic models focusing on information asymmetries and misaligned incentives. As mentioned, the models usually exclude uncertainty, honest mistakes, and genuine differences of opinion.”

He feels that these essential routines around decision making are the pragmatic result of practical experimentation, not the implementation of theories from the dominant economic models.

And in these routines, where leaders surface and discuss their doubts about the future, this experience has shown that building cases with stories is quite effective.

On Narrative

Like Kay and King (and their endorsement of reference narratives), Bhide believes that successful entrepreneurs must be master story-tellers.

Plausible is the key word here - it’s the burden of proof we place on the evidence, and the aim of the connected causality we weave into the stories. But to consistently drive discourse, we need routines that produce good narratives: “Routines evaluate and help construct imaginative yet plausible interpretations (“meanings”) of what is, was, and could be.”

The narratives are used to grab and focus attention (stories > spreadsheets), drive alignment (via justificatory discourse, dialog, disagreements, discussion), and build a shared understanding around the path and why it’s being taken.

Bhide took a fun detour into Jerome Bruner’s two modes of narrative, from “The Narrative Construction of Reality” (1991):

• Logico-scientific mode: idealizes a “formal, mathematical system of description and explanation.” It “deals in general causes”, seeking to transcend the particular by “higher and higher reaching for abstraction.

• Imaginative narrative mode: leads to “good stories, gripping drama, believable (though not necessarily ‘true’) historical accounts. It deals in human or human-like intention and action,” and the “particulars of experience” that it locates in “time and place.”

This really struck a chord with me. My engineering coursework drilled into me the importance of high-quality, “logico-scientific mode” narratives, to build convincing causal cases. The more math you can use, the better! But I also noticed over my career that others seemed to have a gift for storytelling, and could wield influence on decisions without all the math. This felt like a shortcut to me: where was the justification? Where was the evidence? Where was the logic? What, no spreadsheet? It’s just a story!

Once you acknowledge that the key to making decisions under uncertainty is to:

1. Enlist others, 

2. Build a case that is plausible, 

3. Support the case with evidence - yes - but also with a human story that moves from a beginning (the past), to a middle (now), and to an end (into an imagined future), and to… 

4. Talk about the doubts that impede conviction,

…then you begin to see the importance of both modes.

Bhide points to a William James quote (which predates Bruner’s work) that also suggests the two modes: (As a William James fan, I just had to include it here too…)

So while abstract thinking, categorization, generalization, and deduction all have a role to play in building cases under uncertainty, make sure you can sum it up in a compelling story. After all, the future is fiction (for now), right?

Bhide goes further on how and when to use the different modes:

“Notwithstanding its impressive contributions to science and technology, abstract deduction is most dependable for entities governed by the laws of nature, or in the case of software, logical rules. Predicting and managing human conduct - to whatever degree possible - still relies mainly on contextual inference. Similarly, discourse about subtle contextual inference often demands natural metaphorical language rather than abstract mathematical symbols. Using algebra instead of words would not make the arguments Supreme Court justices hear any stronger and the opinions they write clearer or more logical.”

And surprisingly, he even makes a case that those BS numbers in your forecasted demand spreadsheet for annual planning (you know the ones…) are performing a valuable role - in making your narrative more plausible!

“Using imagined details does not require gullible listeners. Assumptions can provide helpful evidence about promoters: can they construct coherent models and defend their assumptions?”

Numbers aside, if a listener sees the logic in your spreadsheet, your overall story might be seen as plausible (even if the number you landed on for your EV adjusted 3-year NPV is not seen as plausible…).

Here’s additional, fundamental guidance on what makes a good narrative, from Bruner’s “Minding The Law” (2000):

• Narrative needs “a cast of human-like characters, beings capable of willing their own actions, forming intentions, holding beliefs, having feelings.

• Narrative needs ”a plot with a beginning, a middle, and an end, in which particular characters are involved in particular events.”

• Plots require:

  • “An initial steady state grounded in the legitimate ordinariness of things,”

  • “That get disrupted by a Trouble consisting of circumstances attributable to human agency or susceptible to change by human intervention,”

  • “In turn evoking efforts at redress or transformation, which succeed or fail, so that,”

  • “The old steady state is restored or a new (transformed) steady state is created, and”

  • “The story concludes… through some coda - say, for example …with a moral of the story.”

Nothing earth-shattering here, just a good reminder that - whether you acknowledge it or not - any time you present or propose things at work, you are composing a narrative. Might as well punch it up (with this guidance) to help pierce any fog of uncertainty in your context.

Bhide struggled to frame a typical entrepreneur’s pitch deck as a prototypical narrative, though. Why? Because pitch decks rarely highlight any Trouble on the expected journey! When your aim is to convince others that you are on (and can stay on) the right track, then the classic narrative form might not fit: 

“Narrative is a recounting of human plans gone off the track, expectations gone awry. It is a way to domesticate human error and surprise.”

You can argue that there is no shape or arc to the story in a pitch deck, just a trouble-free straight-line to profits.

Instead, Bhide thinks entrepreneurs should use narratives in supporting roles, to: 

• Tell origin stories of the company 

• Tell stories about overcoming unforeseen challenges 

• Tell stories about incidents that shaped company culture

• Support sense-making, through stories that become part of the “accepted canon”

• Help shape common language for everyday discourse, via metaphor and figures of speech

• Influence collective norms, with pointed messages or morals of the stories, which are useful for “increasing the potential for agreement-producing dialogue”

Wrapping Up

We’ve covered a lot of ground here. Amar Bhide’s “Uncertainty and Enterprise” is a comprehensive look at how uncertainty has influenced economic thinking over the last century. But it offers more than just a historical review.

By reframing uncertainty as a mental state of doubt, he deconstructs the mainstream model for uncertainty - in terms of subjective (personal) probabilities - into something less quantitative (i.e. degrees of doubt instead of probability) while retaining the emphasis on personal or individual perspective. 

It also opens a door to borrow ideas from how doubts are addressed in fields outside of business, like in legal settings and even in literature. This produced the insight that it is our organizational routines - those that foster group dialog - that help us make sense of things when uncertainty clouds our decision making.

And he reinforces what we heard from Kay and King about the power of story-telling (i.e. reference narratives), written in the “imaginative narrative” mode, to communicate our choices in situations when the evidence (i.e. facts, statistics, or precedent) can’t point the way.

Through this, he adds a crucial layer to our portraits of good entrepreneurs, and leaders, who find the conviction to overcome doubts, and effectively use narrative to communicate their judgment. According to Bhide, this is the path to profit in a modern world dominated by uncertainty.