John L Jerz Website II Copyright (c) 2014

Organizational Improvisation and Organizational Memory (Moorman, Miner, 1998)

Christine Moorman, Anne S. Miner

Academy of Management Review, 1998, Vol. 23, No. 4, p.698-723

JLJ - this paper seems to be the deepest and most thorough look at the concept of improvisation that I have found.

By the way, I would define improvisation as:

1. urgent-composed action
2. that emerges from the needs, dispositions, and the capabilities of the present
3. through a subconscious scheme of reading and reacting
4. to subtle changes in rich information flows and perceived critical connections
5. among both real objects and imagined "aura".

This definition, interestingly, is derived from my experience driving through Washington DC traffic, and putting together ballroom dance patterns in a social setting. In my opinion, all five are required for improvisation. I now expand on the definition:

1. If you are not composing in the present, and composing "urgently", you are not improvising. We might supply this urgency by pressuring ourselves to produce something in a timely fashion "let's see if I can draw a quick sketch here..." - not all urgency has an external source. A professor standing before a class, having reviewed the material, constructed a lesson plan, and prepared notes, begins to improvise a lecture.
2. The present situation you are in, whatever it is - a concert hall, a classroom, a dance floor or a congested highway - exerts a pressure for a certain kind of response/performance, and consequences exist - either reduced social/peer approval, losing one's job or even a harmful accident - for failing to "perform". Our response will be nothing more or nothing less than how we are disposed to respond, given the exact conditions we face, the perceived needs of our situation, and our capabilities to meet those needs.
3. Improvising has a subconscious side where the improvisor might not be consciously aware of anything more than "I have to come up with something here", and having a practiced capability to respond - including a well-stocked "bag of tricks". A professor might prepare for a lecture by making notes and slides, then re-reading the chapters before class. We might begin the scheme by looking to this bag of tricks, prompted by the richly detailed information flows of the present.
4. Monitoring of changes in rich information flows is critical for improvising - a sharp prompt of a change might in effect trip some kind of unknown internal programming - a disposition - to respond in a creative way. Bored-looking students might prompt a professor to simplify the material or return to basic concepts.
5. We might perceive this rich information flow and connections between objects that do not (yet) exist - for example, we avoid the right lane of a highway because we might eventually come up behind a slow-moving vehicle, a subconscious re-remembering of a dance pattern by peripherally becoming aware of another dance couple performing a pattern, or even the perception of being in the left lane of a major highway and imagining that it is only a matter of time before a fast vehicle will appear behind us and hint that it wants us to move out of the way. In this case, we improvise a way around or behind a truck because it is deemed only a "matter of time" before we are pressured to either speed up or move out of the way.

Perhaps a trance-like state is required to improvise, because we are "perceiving" and "reacting" to "things", some of which are not truly "real". Unbelievably, we do not question this trance-like state as being in any way unusual. As humans, we proceed through life from one trance-like state to another, substituting one set of visions and imagined objects for another, deceiving ourselves in pre-planned and common ways.

As a fish ignores the water it swims in, so we ignore the unreality of the everyday and very unreal that we create for ourselves, sustain and co-exist with. Perhaps, shockingly, we react more to the unreal than the real.

But enough of what I feel. Let's look at Moorman and Miner's concepts and ideas.
 

p.698 Observers have long noted the presence and promise of improvisation in the arts, teaching, therapy, and athletics.

p.698 we draw on prior work to define improvisation as the degree to which composition and execution converge in time. Therefore, the more proximate the design and implementation of an activity in time, the more that activity is improvisational. This view of improvisation is primarily temporal, with a focus on the degree of simultaneity of composition and implementation.

p.699 organizations could benefit from deliberate efforts to accumulate memory that will permit them to improvise more effectively over time

p.699 we suggest that the notion of improvisation arises in varied contexts and that the term improvisation has been defined in varied ways

p.702 [Bjurwill (1993)] "Reading and reacting in parallel" (p. 1383); "Dual tasks" (p. 1384); "Perception-in-action" and "thinking-in-action" (p. 1386)

p.702 In improvisation the time gap between these events narrows so that in the limit composition converges with execution. The more improvisational an act, the narrower the time gap between composing and performing, designing and producing, or conceptualizing and implementing.

p.702 "Composition" implies that the improvisational activity involves some degree of innovation, because it goes beyond automatically repeating a pre-existing routine (Crossan & Sorrenti, 1997; Weick, 1996).

p.702 improvisation typically has a referent or "an underlying formal scheme or guiding image" (Pressing, 1984: 346). Yet, the order is not fully predesigned and, in that sense, is partial.

p.704 How, then, does collective improvisation occur? One of the most common metaphors for group improvisation is the notion of "conversation," in which at least two agents interact around a theme or referent to compose while executing. Using this metaphor, we argue that collective improvisation may be produced by the joint activities of individuals, who are themselves improvising. The joint action of the individuals produces a "system" that we label a collective improvisation. Theatrical improvisation offers an obvious example here, when the ultimate shape of a scene arises not from any prior plan but from what unfolds after the first actor generates lines and movement, a second responds to that, and the group continues to interact (Crossan & Sorrenti, 1997; Mangham, 1986; Spolin, 1963).

p.704 We suggest that any entity that can reasonably be thought of as planning or executing action can also be thought of as improvising.

p.705 Bricolage is defined as "making do with the materials at hand" (Levi-Strauss, 1967: 17)... we believe that the more improvisational an act, the more likely bricolage is to occur, because there is less time to obtain appropriate resources in advance. Finally, we argue that being skillful at bricolage may actually help produce valued improvisation.

p.705 Crossan and Sorrenti define intuition as operating when "choices [are] made without formal analysis" (1997: 3) and describe it as central to improvisation (Chase, 1988).

p.705 we suggest that adaptation involves adjustment of a system to external conditions (Campbell, 1989; Stein, 1989)... Adaptation can be achieved when an organization plans in advance to adapt to change. An organization can also adapt by making contingency plans for different competitive situations that may unfold. Finally, an organization can adapt by deploying standard response routines... not only are there many forms of adaptation not involving improvisation as listed above, but even more important, the construct of improvisation does not imply that all improvisation is adaptive.

p.705 If an organization improvises, assesses outcomes, and then acts again, this process can be seen as trial-and-error learning.

p.706 Organizational scholars writing on improvisation typically have emphasized instrumental outcomes... Two specific criteria appear to dominate such outcomes. First, does the improvisation solve a problem? ...Second, does improvisation permit the organization to harvest unanticipated opportunities?

p.707 improvisation... three instrumental outcomes... First, action can vary in the degree to which it displays coherence... the degree to which an action displays internal fit... and external fit... Second... improvisation can produce varied levels of novelty in action... third, speed refers to the time to plan and execute an action.... Others have suggested that improvisation may speed up action because it supplants lengthy planning

p.708 A key characteristic of procedural memory is that it becomes automatic or accessible unconsciously. Often referred to as "motor memory" (Pressing, 1988), procedural memory includes the skills needed to ride a bike or use a typewriter (Cohen, 1991; Pressing, 1984; Singley & Anderson, 1989). Procedural memory, therefore, often represents tacit knowledge for individuals and organizations (Cohen, 1991; Cohen & Bacdayan, 1994; Nonanka, 1990; Winter, 1987).

p.709 Proposition 1: The greater the procedural memory level, the greater the likelihood that improvisation will produce coherent action.

In addition to increasing the likelihood that improvisation will produce coherent action, procedural memory is also likely to improve the speed of improvisation. Increased speed is due to the automatic or tacit quality of procedural memory, thereby producing an "economy of action" (Pressing, 1984: 335).

p.709 Proposition 2: The greater the procedural memory level, the greater the likelihood that improvisation will produce speedy action.

A high level of procedural memory can also have a third and contrasting effect on improvisation, which is to constrain novelty.

p.710 Proposition 3: The greater the procedural memory level, the greater the likelihood that improvisation will produce action low in novelty.

p.711 In some settings production professionals believe engineers and others with strong theoretical knowledge create problems by searching through far too many forms of abstract knowledge, instead of drawing on heuristics that can be deployed rapidly (Dean & Susman, 1989).

p.712 declarative and procedural memory can be viewed as complementary competencies that offset one another's weaknesses. As a result, their joint deployment may be especially effective in ensuring that improvisation results in coherent, novel, and speedy action. Bjurwill (1993), in the sports literature, describes the criticality of both mental skills (declarative memory) and motor skills (procedural memory) for thinking in action or improvisation.

p.712 the degree to which improvisation produces novel action may depend on whether the improviser is able to use declarative memory to make creative use of procedural memory.

p.712 we suggest that the degree to which improvisation positively impacts speed may depend on whether the improviser has developed procedural skills that allow for rapid access to declarative memory stores. Such skills require the ability to find and incorporate declarative memory into action as improvisation occurs.

p.713 Observation of improvisational actions may serve as experiments that shape future behavior and the memory of the actors.

p.713-714 Theories of organizational evolution and and trial-and-error learning emphasize that random variations in organizational activity may produce unexpected benefits. Firms then observe the beneficial consequences of these random variations, and they repeat and even institutionalize these accidentally discovered actions over time so that they become part of the organization's memory... this can constitute a process of (unplanned) adaptation... In these models the initial activity - later repeated - may be nonrandom, possibly resulting from efforts to solve an immediate problem.

p.714 Proposition 8: Improvisational actions can serve as unplanned experiments that generate changes in an organization's procedural and declarative memory.

p.714 we suggest that it is possible for an effective improviser to acquire - through practice - a general skill or competency in improvisation itself. This skill, while procedural, represents a metaroutine, as opposed to a single routine that the improviser can access during improvisation... this skill can be learned... it can, in turn, influence a variety of improvisational outcomes

p.714 Brown and Eisenhardt (1995)... found that new product development... relied on real-time interaction and a targeted structure... the real-time interaction... increases the amount of knowledge available for recombination in the firm. Brown and Eisenhardt (1995) also argue that structured managerial roles and clear project priorities are essential to effective improvisation. In our framework this structure is likely to evoke routines (procedural memory) at the appropriate time and place.

p.715 the appropriate focus should be on the simultaneity of design and action - not on whether planning should or can be eliminated from organizations.

p.716 the propositions imply that improvisation may be ineffective at best, and possibly harmful, unless an organization has a rich repertoire of procedural memory, in the form of organizational routines, and a deep reserve of declarative memory, in the form of more abstract or theoretical information. Although not startling, these ideas imply a fairly unusual combination of organizational memory that must be developed and deployed for improvisation to represent a fruitful tool in organizational effectiveness. [JLJ - Yes, you can end up looking foolish, disconnected from reality, or inattentive to important concerns unless adequate guiding frameworks are ready-on-tap, practiced, and poised to be activated by the richly-sensed, real-time information of the present and immediate past.]

p.716 we suggest that both procedural and declarative memory can reside in social structures and practices (such as group values or routines) and in material structures and practices (such as in the placement of a building or maintenance routines for the building [JLJ - echoes of Bourdieu's habitus]

p.719 our framework suggests that procedural memory should enhance improvisational effectiveness and speed, while reducing its novelty. Declarative memory, however, should enhance improvisational effectiveness and novelty, while reducing its speed. Given these tradeoffs, we proposed that the presence of both procedural and declarative memory would be especially likely to produce valuable improvisation