Which Entrepreneurship Paradigm? Exploring the Epistemic ... · explored the topic of “keystone...
Transcript of Which Entrepreneurship Paradigm? Exploring the Epistemic ... · explored the topic of “keystone...
Paper to be presented at the
35th DRUID Celebration Conference 2013, Barcelona, Spain, June 17-19
Which Entrepreneurship Paradigm? Exploring the Epistemic Properties of
Keystone RulesFabrice L. Cavarretta
ESSEC Business SchoolManagement
Nathan R. FurrBrigham Young University
AbstractBoundedly rational actors cannot instantiate the full spectrum of theories in the world. Although management researchhas identified ?simple rules? as guiding specific types of action, other research fields suggests the existence of higherorder ?keystone rules? that guide the selection of simple rules themselves. We conduct a grounded exploration ofentrepreneurs? keystone rules and attempt to consider them as epistemic objects. Although they do not match scientifictheory on all dimensions, they cannot simply be rejected based and provide an insight into the possible epistemic bodiesthat might practitioner actions.
Jelcodes:M00,-
1/35
Which Entrepreneurship Paradigm?
Exploring the Epistemic Properties of Keystone Rules
SHORT ABSTRACT
Boundedly rational actors cannot instantiate the full spectrum of theories in the world. Although management research has identified “simple rules” as guiding specific types of action, other research fields suggests the existence of higher order “keystone rules” that guide the selection of simple rules themselves. We conduct a grounded exploration of entrepreneurs’ keystone rules and attempt to consider them as epistemic objects. Although they do not match scientific theory on all dimensions, they cannot simply be rejected based and provide an insight into the possible epistemic bodies that might practitioner actions.
LONG ABSTRACT
Practitioners’ theories of action mostly appear to scholars as naïve--the results of simplistic social construction or the product of various biases. At the same time, specific heuristics and simple rules are increasingly seen as functional antecedents of managerial action. Yet, given a highly complex world, increasing rule sets, and boundedly rational individuals, actors cannot instantiate the full spectrum of theories of the world, however simple they may be. Conceptually, the elaboration of a theory of action amounts to selecting an optimal set of a few good rules out of a large set of possible rules, which constitutes a highly combinatorial problem with a local optimum solution at best, such as we observe in genetic algorithms. In practice, we observe that many fields, such as medicine and military science recognize the existence of “keystone rules” that guide the selection and use of heuristics themselves. Although little formal research has explored the topic of “keystone rules,” casual observation in entrepreneurship suggest that much of practice, teaching, and research are guided by a few overarching, perhaps un-examined, set of rules that guide the selection and execution of more commonly studied simple rules. To explore this possibility we conduct a grounded exploration of entrepreneurs “theories of action” and attempt to consider them as epistemic objects. We focus in particular on the case when actors express a top-level overarching theory of their action. Employing a broad sample of primary and secondary data, we use categorization analysis to explore the existence of keystone rules that appears overwhelmingly in the sample. An epistemic analysis of these keystone rules suggests that their epistemic validity cannot be rejected, even though they do not match scientific theory in various dimensions. We propose to label such simple rules as keystone rules because of their architectural role on the emergence of action, and the possibility that such clusters may signal a paradigmatic structure in the epistemology of action. Finally, we relate the emergent set to the effectuation principles. Even though the paradigms do not perfectly matched, they have strong similarity in their contrast to a causation paradigm, hinting that reasoning at paradigmatic level may constitute fruitful future direction for research.
2/35
“I think the basics of being an entrepreneur are still very much the same as when I started
Virgin 40 years ago. Entrepreneurs should be open-minded, prepared to listen, and also expect
to experience rejection and setbacks. It helps to know your limitations and surround yourself
with people who have skills you may lack. Also, a vital part of being entrepreneurial is being
decisive and being prepared to take risks - don’t be afraid to follow your guts”
Richard Branson, 2010
INTRODUCTION
The division of epistemic labor in management studies has evolved towards a self-evident
truce where scholars produce theories according to a scientific method, and practitioners
practice. As for practitioners’ expressed theories, in particular their theories of action, they
appear mostly as objects of epistemological analysis than being attributed much epistemic value
per se.
On the one hand, behavioral scholars, rightfully expecting limited rationality, approach
actors’ theories of action as being sub-optimal and, not surprisingly, have demonstrated a wealth
of biases in actors’ decision making processes (Cyert & March, 1963 [1992]; Denrell & March,
2001; Levinthal & March, 1993). On the other hand, social constructionists shun notions of
optimality of decision-making but focus on the various social factors and processes influencing
the construction of mental representations. Hence, they interpret theories of action as cultural and
institutional elaborations (Weick, 1979), a perspective that makes it difficult to attribute them
much epistemic qualities.
Yet, a few alternative perspectives have started to challenge the idea that practitioners’
theories of action have little epistemic qualities. Heuristics have been demonstrated to carry
significant accuracy in psychology literature (Todd & Gigerenzer, 2003), an idea reflected in
3/35
strategy scholarship where “simple rules” have been demonstrated to drive strategy (Bingham,
Eisenhardt, & Furr, 2007; Miller, 1993), and a foundational element of the cognitive architecture
that drives managerial action (Gavetti & Rivkin, 2007). Studies have explored their emergence,
for instance Bingham and Eisenhardt considering mid-range rules of the type “enter countries
with lots of pharma activity” (2011:1444), or have explicitly positioned simple rules as
derivative of overarching values and representations (Gavetti & Rivkin, 2007:432).
However, although cognitive processes can occur in hierarchically nested structures (e.g.,
Argyris & Schön, 1978; Gavetti, 2012), little research has explored the possibility that some
rules could play a more central architectural role than other rules. Contrast this with the belief
expressed in the opening citation by Richard Branson, which exemplifies a widespread empirical
phenomena occurring in business practice in general, and in particular in entrepreneurship
context: that a limited set of rules shape the selection of other rules and determine most of the
variance in outcomes. In many parallel fields, such as military science, practice has recognized a
clear set of keystone rules to guide action, and while management fields, such as
entrepreneurship, may not appear to have such rules, beneath the specific practitioner voices,
there appears a consistent effort to identify a set of keystone rules, evidenced by the vigorous
Nonetheless, below the many specific practitioner voices, there appears a consistent effort to
exchanges of advice and the robust press and publishing industries where authors, journalist and
practitioners dialog in attempt to identify the “key” insights of a focal activity. Indeed, the few
rules stated by Branson above suggest a top-level set of rules, at least in the minds of actors,
which could holistically shape the entrepreneurial process. These rather common expressions
among entrepreneurs raises the possibility that rules have greater structure than previously
appreciated, and in particular there could exist, among all possible rules, a limited, guiding rule
4/35
set that would have a disproportionate impact because they orient action and also learning and
sense making, hence have architectural consequences. This led us to label them keystone rules.
This paper explores such logic by taking the assumption that, for individual managerial
actors in the management science field, such as entrepreneurs, their theory of the world will rest
on a few rules that are somewhat permanent and operating at the highest-level of cognitive
process. Furthermore, this bundle would be small per the most fundamental interpretation of
bounded rationality (Simon, 1947 [1997]). This does not assume that cognition is actually so
limited, but rather that the set of keystone rules will orient various mechanisms such as learning
and self-selection, and that actors will instantiate a fundamentally greater range of cognitions
from those processes and from exposure to the environment. We therefore aim to explore how, in
such tiered cognitive structure, a few particular simple rules could play a particular architectural
role, and what could be the epistemic properties of such bundle, if they exist at all
In a first section, we demonstrate the theoretical possibility of “keystone rules” and that the
selection of bundle of rules constitutes a hard combinatorial problem (e.g., Rivkin, 2000) that can
only receive a local optimum solution, such as genetic algorithms. Furthermore, we identify that
the entrepreneurial context constitutes a remarkable natural setting for such genetic algorithms,
which leads to a grounded theory exploration of such keystone rules.
Empirically, we conducted a grounded theory empirical exploration of entrepreneurs
expressing their “theories of action” (Argyris & Schön, 1978), in particular in form of grand
theory, when those actors expressed that a few principles are fundamental to their practice. We
gathered and selected archival interviews of entrepreneurs, selecting on instances where they
express such grand theory. We subsequently conducted primary interviews to explore further the
nature of the theories that entrepreneurs expressed, the weight that these actors assign to their
5/35
theory, and their perceived rationale. The analysis consists of a two-level categorization, leading
to identify a corpus of theories of action. This first emergent phase then allows identifying a
kernel of theories of action that constitute a candidate for keystone rules in that context.
We then conducted an epistemological analysis of these keystone rules. We identify their
negative epistemic properties: fuzziness, inconsistency, counter-factual. Simultaneously, we find
these to have positive epistemic properties: they match cognitive capabilities, are polymorphic
and self-fulfilling. These findings contribute to the emerging literature on simple rules and
heuristics by suggesting the importance of simple rules but adding the importance of rule
architecture, particularly keystone rules, that shape other rules. Although these keystone rules
may appear simplistic, even biased, when viewed through the lens of academic study, their
prominence in practice suggests the need to more robustly identify, observe, and perhaps
influence practitioners’ keystone rules. Therefore, while research has suggested the importance
of simple rules in action (Bingham et al., 2007; Davis, Eisenhardt, & Bingham, 2009), this study
suggests the opportunity to study more robustly the character and architecture of these rules.
Furthermore, this study contributes to the literature on performativity and social calculation
(Callon & Muniesa, 2005)by suggesting the specific importance of practitioner keystone rules—
rather than being objects of dysfunctional naivety, keystone rules are potentially powerful
cognitive objectives deserving robust study.
THE HARD PROBLEM OF DETERMINING AN OPTIMAL SET OF
KEYSTONE RULES
To determine the bundle of rules with maximum epistemological quality, we first review the
motivation to assume actors view their world through a limited set of rules. Then, we explore
why the selection of such bundle constitutes a hard combinatorial problem, and how the chatter
6/35
among entrepreneurs regarding grand theory of action might constitute a relatively efficient
approach to developing an “algorithm.”
Assuming the Existence of a Keystone Rules
The core idea of bounded rationality is that humans have limited cognitive resources, such as
memory, process, attention, etc. (Simon, 1947 [1997]), hence putting constraints on the number
of rules to which actors can be exposed, memorize, pay attention, and process. One can also
interpret this through an attention-based view of organizational life, whereby the limited
attentional capabilities of actors drive outcomes (Ocasio, 1997, 2011).
Such constraints are acute for all actors but may be even more acute for entrepreneurs, a class
of managerial actors that is most identified with individual action early in the life of a new
venture, without the help of an established organization to process and manage information.
This constraint may be compounded further in the entrepreneurial context since many individuals
engage in entrepreneurial activity without having been exposed to the large and structured body
of knowledge delivered in business schools.
Overall, when it comes to their theory of management, entrepreneurs are likely to have to
elaborate a small set of simple rules. For the purpose of the rest of the reasoning, we will stylize
the above constraint as putting a hard limit on the number of theory entrepreneurs can hold. We
could further elaborate on this issue, as the nature and consequences of bounded rationality has
already motivated most of managerial sciences, among others in the Carnegie tradition, the
behavioral perspective, the cognition literature, etc. Establishing such stylization is not the
purpose of the current study, and we rather state here clearly a simple and crucial assumption
onto which this study rest: actors view the(ir) world through a short bounded set of simple rules.
This assumption does not require that cognitions are actually so limited, but rather that
7/35
because actors face limitations, there exists the possibility that actors develop not only simple
rules, but also develop a set of higher-order keystone rules, that orient various mechanisms such
as individual learning, and the self-selection of actors into various activities. On an ongoing
basis, actors can instantiate a fundamentally large range of cognitions even when employing
keystone rules, hence our assumption is not incompatible with rich cognitive and social
processes such as (vicarious) learning, social construction, communication, etc.
The above assumption does not qualify the actual size of the rule bundle. However, as we
observe in the empirical portion, the reasoning for keystone rules—paradoxically—holds all the
more if we argue that entrepreneurs can host a large cognition set, since the only way a
boundedly rational individual can engage in such rich cognition is through a set of keystone rules
that organize and direct the more rich cognitive practices. Furthermore, as we observe in the
empirics, in practice, entrepreneurs state keystone rule bundles that range from one to a dozen
rules of action.
Finally, the reader might wonder why we assume the existence of such bundle and whether
the demonstration of such existence should rather be the purpose of the paper. However, we
insist that we leave aside, on purpose, the technicalities of a cognition study (Walsh, 1995) that
would aim to robustly demonstrate the robust existence of specific keystone rules (e.g. name the
ten keystone rules held by entrepreneurs), to instead inquire what would be the properties of
such keystone rule bundles.
The Hard Problem of Determining an Optimal Keystone Rules Set
The various theories from which the entrepreneurs can draw to build a bundle of rules is a
large. In particular, it includes the theories already available and verified in the management
science literature (the “traditional” epistemology). In order of magnitude, it can be estimated by
8/35
multiplying the number of hypotheses produced by all journals for all years for decades,
suggesting thousands if not tens of thousands of rules. In addition, the explorable set contains
also all rules used in pedagogy or in practice that are not yet validated in traditional
epistemology. Finally, the explorable set contains any other rules that may emerge from
experience or social construction, either individually or collectively. Already, the set of possible
theories to draw from is tremendously large, at least compared to the bundle an individual
entrepreneur can cognitively accommodate.
This discrepancy, a fundamental interpretation of Simon’s bounded rationality idea,
constitutes the major constraint on the determination of the optimal bundle of rules. Let us first
notice that science in our positivist tradition amounts to state independent relationships (e.g. ,
X=>Y) and then verify empirically whether or not the hypothesis holds. In this approach, the
determination of truth amounts to a binary outcome of an individual test. Collectively, the
traditional epistemology amounts to accumulating the sequential verifications of unitary
hypotheses.
However, once the total set of validated hypotheses have been established, little is known
about which subset of hypotheses would be cognitively efficient for actors in practice, assuming
one is cognitively constrained. The reason is that by contrast to individual hypothesis testing
(e.g., is hypothesis A true or not?), selecting an optimal subset amounts to a combinatorial
problem (e.g., how to select an optimal A, B, C among a large set of theories?) that has long
been identified as intractable (e.g. Rivkin, 2000).
Let us qualify what makes this a hard problem: evaluating the validity of a limited size
bundle (e.g. 3) to be picked out of a large set of validated theories (e.g. out of 10.000) implies
that it is better than any other combination. In practice, the bundle would have to be compared to
9/35
an outstandingly large alternative set (e.g., for 3 out of 10.000, the alternatives amounts to
roughly 1012, a thousand billions) making it hard since it implies non-human time and scales.
The only case where such hard problem would be solvable happens if the rules were to be
disconnected, non-interdependent. In which case, one might imagine actors could sort the rules
depending on their effect size (Combs, 2010) and retain only the most important rules. Beyond
the naivety of this approach, it would nevertheless be irrelevant since interdependence abounds
in management science, in which case, the linear approach does not apply (Rivkin & Siggelkow,
2003). Overall, it appears that an optimal bundle of theory can neither be exhibited nor validated
by a positivist approach.
Although forming a simplified bundle of heuristics may be difficult using the traditional
hypothesis testing framework, sciences that have been confronted with similar combinationally
difficult problems have explored alternative solutions, for example, the genetic algorithm
(Holland, 1975). In practice, a genetic algorithm requires a massively large group of actors
experimenting with various combinations of the rules, exchanging those rules among themselves,
and assigning a reproductive advantage to individuals that over-perform. However, under these
constraints, genetic algorithms have been demonstrated to converge efficiently towards local
optimum, given that pure optimality remains out of reach in most such hard problems. Not
surprisingly, genetic algorithm have already been invoked in organizational studies as a solution
to select sets of routines (Bruderer & Singh, 1996) or sets of strategy rules (Rivkin & Siggelkow,
2003).
As it happens, the context of entrepreneurship provides exactly the natural experiment
conditions to develop a “genetic algorithm.” The robust expression of grand theories of action
(by direct contact or through the press) is the exchange mechanism; the breadth and speed of
10/35
entrepreneurs experimenting provides variations; and the selective pressures of entrepreneurship
occur through death of firms, exit of entrepreneurs; finally, reproductive advantage condition
occurs since voice is given in priority to successful entrepreneurs (Aldrich, 1999:chap. 4). The
ecology of memes perspective (Weeks & Galunic, 2003) concurs with this interpretation since it
identifies such types of expressed belief as being subjected to evolutionary processes which
would occur jointly to the evolution of entrepreneurs. Overall, one can interpret the
entrepreneurship context as constituting a natural genetic algorithm where both individual and
bundle of cognitions are co-selected. Therefore, the emergence of a small bundle of rules that are
commonly cited by entrepreneurs suggests that such rules adaptive, at least relative to the
constraints that entrepreneurs have in carrying and exchanging rules.
This formal reasoning suggests that the rules that emerge from entrepreneurs exchanging their
grand theory would constitute a local optimal bundle of theories. This reasoning does validate
per se the bundle as being optimal, indeed, the demonstration above state clearly that, as it is
framed, no one might be able to validate optimality. To be clear, the positivistic hypothesis
testing approach, applied to those theories of action (e.g., evidence based management in Pfeffer
& Sutton, 2006), can at best determine which of a limited number of sets (e.g., if two: A, B, C vs.
D, E, F) is better, but once this is settled, is completely mute about whether this is even a local
optimum. However, this suggests how such keystone rules might fill the theoretical gap between
the rational decision and social construction approach. Relative to rational decision approaches
that viewed entrepreneurs theories as systematically biased, the bounded rationality constraint is
now externalized and made a defining constraint of theory construction. The emergence of
shared belief that was considered by social constructionist as an institution devoid of optimality
is now made a mechanism to resolve a hard social problem: the identification of entrepreneurship
11/35
theory that fits how entrepreneurs actually conceptualize and discuss their practice.
EMPIRICS
Context
We chose to focus our exploration of grand theories of actions in the entrepreneurial setting ,
because entrepreneurs provide a particularly favorable context to observe as clean a genetic
algorithm as possible. First, the role of entrepreneurs, by contrast to various other managerial
roles such the role of managers in existing firms, is to exhibit a great level of individual agency.
Second, because of acute evolutionary processes in entrepreneurial contexts (Aldrich, 1999), the
entrepreneurial setting provides a rich context in which to study keystone rules. Finally, because
society gives voice to entrepreneurs, particularly successful entrepreneurs, a clear dissemination
and selection mechanism may drive the development of keystone rules.
Because we care about the possibility that the natural setting may provide a relatively
efficient genetic algorithm, communication is of the utmost importance, hence our focus falls on
the grand theories of action as they are explicitly expressed by entrepreneurs. We do not ignore
that expression and actual usage of rules might differ1, our study focus on rules as they are
expressed since rules that are private to individual cannot contribute to any shared algorithm.
There is also the possibility that actors observing other actors could lead to vicarious learning,
even though their rules are not explicitly expressed. This situation is less likely to lead to
transmission in the context of entrepreneurship due to atomization of work habitus. Finally, a
rule may be never expressed but clearly at work for social desirability reasons (e.g., selfishness,
greed). We will discuss this concern in the section below, since the presence—or absence—of
socially stigmatized rules might constitute an epistemic characteristic of the set of rules.
1 The discussion section will revisit the tradeoff between studying expressed – i.e., exchanged – rules vs. studying rules in use [to be written]
12/35
Sources of Data
To explore the possibility of keystone rules, we conducted a qualitative study, first gathering
a set of archival documents and then conducting primary interviews where such beliefs are
expressed by a sample of US and French entrepreneurs.
Grand Theories in the Wild: Archival Exploration - We explored a large sample of
entrepreneurs’ interviews in the press, searching by keywords (e.g. “entrepreneurs”,
“interview”), and by sources (e.g. MIT Entrepreneurship Review). The research assistants
conducting the search were instructed to retain paper where spontaneous expression of grand
theory appeared, and to eliminate cases where grand theorizing is suggested in the question of
interviewer. Out of that search, we selected a representative sample of 15 interviews to be
included in the qualitative analysis.
Semi-structured Interviews - We then conducted primary interviews (15) to directly probe the
existence of grand theory and explore the logic as expressed by entrepreneurs. An interview
protocol, of semi-structured type, which is provided in Appendix A. It was established based on
the analysis of the archival exploration.
In addition, even though these events are not directly included in the analyzed data, the
researchers associated with the project have conducted roughly 100 unstructured interviews with
entrepreneurs around those issues.
Sampling
Such qualitative data collection entails a difficult boundary selection problem. Taking too
large of a community may create too much variance. On the one hand, sampling entrepreneurs
across countries (e.g., US vs. India) and across types of entrepreneurship (fundraising in Silicon
Valley vs. mom and pop survival business) appeared overly ambitious given the differences in
13/35
these contexts. On the other hand, narrowing on all dimensions (country, demographic, type of
entrepreneurship, industry, etc) was not appealing nor practical given constraints to
generalizability and since the archival data collection had to be relatively large to ensure finding
spontaneous expression of grand theory.
In the process, the data collection (archival and primary interview) converged onto a
reasonably bounded community of entrepreneurship that is implied in the institutions of
international business schools. These entrepreneurs were mostly well-educated, male,
international (mainly US, France and a few others), had engaged into scalable entrepreneurial
projects (i.e. higher than normal survival probability and with greater ambitions than typical
small business (mom and pop projects: i.e., significant project that would entail significant
investments or revenues)).
Analytic Approach
We performed nested levels of coding, using Atlas TI, a qualitative data analysis software,
allowing in the analysis both openness in initial phase, as well as comprehensiveness and rigor in
the later classification phase. We began by identifying a large breadth of expressed theories at a
first level, using an open coding of basics rules, using in-vivo codes (Strauss, 1987). This first
pass of analysis demonstrates a large spectrum of rules with apparently little similarity across
practitioners. We summarize the pattern of citation of level 1 rules in Table 1.
----- Insert Table 1 roughly here -----
Then, at a second level, we conducted an axial analysis by searching for relationships
between categories in order to assemble them into higher order themes that agregate and
generalize related rules (Yang, 2010). This selection and identification was conducted in a
grounded theory approach, whereby meaningful associations take precedence over the
14/35
temptation to rely on quantitative counts. This process of abstracting rules into higher-level
categorization—which are labeled keystone rules when discussed below—operated both as a tool
for recurring passes of analysis, as well as producing in the end a final categorization once
saturation was obtained. To illustrate how it converged into a selection of keystone rules, we
provide an intermediary grouping of rules (Table 2) that was built by simply aggregating rules by
related content.
----- Insert Table 2 roughly here -----
This intermediary grouping was subjected to additional transformation upon analysis. First,
additional associations were conducted across the groupings (e.g. Irrationality and Autonomy
were consolidated into a meta-rule “be self-determined to the point of irrationality”. Second,
some groupings (e.g., “Not from school”, “Attention, Strength”) were not retained in the final
bundle of keystone rules discussed here, mainly for substantive reasons, even though these
groupings also had less occurrences than all the retained keystone rules. In particular, the
keystone rule “not from school” is not so salient in the responses because the interview context
(university researchers)possibly cued the response rather than this forming a fundamental
keystone rule guiding other rules.
In this process, we also took into account metrics to guide the selection of rules based on
their salience. Table 3 illustrates the basic metrics used when working on the groupings: the
count of all mentions, the count of appearance (i.e. 1 if appears in a text, otherwise 0), and the
appearance as a ratio.
----- Insert Table 3 roughly here -----
Considering the grouping “Attention”, which bundled rules stating what entrepreneurs should
pay attention too, it aggregated “attention to customers” and “attention to employees”. This
15/35
grouping was useful conceptually in the initial phase, but it appeared overly general (it
aggregates any idea of “paying attention to something). Hence, we decided to consider meta-rule
by breaking it back into separate substantive attention (customers vs. employees). Quantitatively,
we put a symbolic and tentative threshold for retaining a meta-rule that it be mentioned at least in
50% or more of the cases, which threshold “neither attention to customers” not “attention to
employees” came close to. Regarding the grouping about “strength”, it was also eliminated
simply because of its very low occurrence ratio. Overall, this process allowed extracting from a
broad set of expressed theories of action, a kernel of meta-rules, summarized in Table 4.
----- Insert Table 4 roughly here -----
FINDINGS
Dominant Meta-Rules
In reviewing the qualitative data, three keystone rules emerge strongly, mentioned in more
than half of the cases, hence constitute a good candidate as keystone rules in this context. These
will be the one for which we will provide further substantive details..
The first most common meta-rule addresses the possibility of failure, by stating that the
entrepreneur should embrace the possibility of failure. It appears as the aggregation of three
relatively distinct sub-rules. The first one suggests that entrepreneurs should expect failure. The
wording often varies, for instance, failures are sometimes evoked as “mistakes”, or the concept
of failure appears at firm level, individual level, or innovation / product level. The second rules
concerns learning from the failure that has been predicted to occur from the first rule, with a
large proportion of actors stating that failure matters in that it is a crucial path to learning. The
third rule states that one must find the strength to try again after failure, both regarding the
humiliation of the first failure, and regarding the possible repeat of such failure.
16/35
The second most common meta-rule concerns the idea that entrepreneurs be undeterred by
negative feedback, while the tension with the necessity to accept and learn from such feedback is
acknowledged. This rule is often complemented by a sub-rule that entrepreneurship requires a
significant amount of irrationality.
The third most common meta-rule concerns the best motivation for entrepreneurship,
associating success to “passion” (i.e., a deep emotional motive quite distinct from expecting
monetary rewards), mainly as a mediator of the persistence into the entrepreneurial effort.
Negative Epistemic properties of keystone rules
Although in this paper we can only begin to identify possible keystone rules, identifying and
expressing such meta-rules raise a wealth of issues for social scientists, of which we identify here
the three main negative characteristics: fuzziness, internal contradiction and externally invalidity.
Fuzzy categories - Categorizing the rules is made somewhat subjective because of semantic
issues. Many of the keystone rule expressions tend to be imprecise, embedded in everyday
language. For instance, the meta-rule regarding failure will sometimes be worded by the word
“failure,” other times by the word “error,” still other times by “mistake,” and so forth. The rule
boundaries also varied based on their function: sometimes they are descriptive, associating a
characteristics with the nature of the entrepreneur; sometimes they are normative, associating the
same characteristics with success; and sometimes they fall into symbolic range (e.g. the “be
foolish” rule by Steve Jobs as an example of a rule falling into our “self-determined to the point
of irrationality” keystone rule category).
Inconsistencies - Inconsistencies and biases have been at the heart of the study of heuristics
(Kahneman & Tversky, 1979). Accordingly, our study of grand theories of action unearths
various inconsistencies, beyond the fuzziness already identified above. The most obvious
17/35
example occurs on the issue of rationality. The rule that entrepreneurs should exhibit some
stubbornness and resistance to outside influences is dominant, yet simultaneously, the rule also
persistently mentions that entrepreneurs should learn from feedback. Most entrepreneurs who
had mentioned “do not listen to advice”, when explicitly asked about it, acknowledge the benefit
of feedback and advice. This creates an interesting contradiction whereby the natural rule
expression exhibits interesting inconsistencies, at least across actors and sometimes at the level
of individuals. Assuming this might signal a case of “ambidexterity” (Tushman & O'Reilly,
1996), these two—apparently contradictory—aspects can be integrated into a single keystone
rule.
Factual Inaccuracies – Some of the interpretation of the rules can make then factually
untrue, or at least controversial or difficult to support. For instance, the idea that entrepreneurs
should be “risk takers” emerges frequently, yet research has demonstrated that entrepreneurs are
indeed more risk averse than non-entrepreneurs (Hongwei & Ruef, 2004). Similarly, rule to be
self-determined to the point of irrationality could be interpreted as a typical case of optimistic
overconfidence, which has been shown to lead to detrimental outcomes (Simon & Shrader,
2012). Finally, even though passion has intuitive appeal, its actual effects have been difficult to
pin down (Chen, Yao, & Kotha, 2009).
However, notice that each rules can lead to various interpretation, which itself can lead to a
large literature with various branches to consider. For each of them, as exemplified above,
probably one could find a contradiction to the general idea. Hence, the validity of the rule in its
general expression probably does not make sense, and many of its derivation could be easily
rejected in a classical scientific epistemology.
18/35
Interesting Epistemic Properties of Keystone Rules: a Difficult to Reject Rationality
Even though properties of keystone rules confirm the general perception that they cannot—a
priori—be attributed validity in the classical scientific sense, our analysis suggests some positive
epistemic properties that might make them useful and adaptive.
Cognitive Dimensioning and Generativeness - Both in the archival data and in our
interviews, actors exhibit the desire to identify and express a small bundle rules that drive their
entrepreneurship effort at a high level. This signal is consistent with our assumption that
boundedly rational individuals form and use a set of keystone rules to guide the many more
actionable heuristics that have been identified by research in specific contexts.
By construction, we have probed the apex of the belief system of entrepreneur. The data
confirms that entrepreneurs spontaneously organize their belief in a tier system, with a high
prevalence of them identifying a small set of belief as being “fundamental”. This tiered schema
has strong analogies to the idea of second-order learning (Argyris & Schön, 1978), whereby
some cognitions would play a disproportionate and generative role towards the rest of the
cognitive structure (Alessi, 1987). Here, if acted upon, the rules in the bundle we identify
(acceptance of failure, role of passion and self-determination up to irrationality) orient the
mechanisms of learning and have significance consequences on entrepreneurs’ performance
trajectories.
Polymorphism – Even in their ambiguities, keystone rules carry a bundle of beliefs. For
instance, the “be prepared to meet failure” theory of action is interpreted variously across actors
but also variously intra-individual. For instance, one actor might mention this idea in a manner
that might appear precise (“Entrepreneurs like risk”) but then moves smoothly around the
spectrum of the meta-rule, and ‘explains’ the previous sentence as meaning that “they should not
19/35
fear failure”, finally stating that the key point is to “be able to recover”. Therefore, what could be
measured as an isolated cognition if trying to capture a narrow idea is actually the source of a
large spectrum of cognitions, which are subsumed in the keystone rule. Hence, such keystone
rules can be viewed less as a bundle of isolated and well-defined theories, but more as exemplars
of a polymorphic and broad idea.
Self-fulfilling - The keystone rules also have the property of being self-fulfilling prophecy—
or as being “performative” (MacKenzie & Millo, 2003). For instance, the statement that
entrepreneurs “do it by passion (not for money)” could be challenged in traditional epistemology.
However, it entails a strong desirability dimension that will affect the dynamic of actors and
beliefs in the field. In particular, since our interviews incorporated actors who are also in position
to judge and select entrepreneurs (i.e., venture capitalist), the keystone rules appear as having an
influence on who is likely to get funded and who is not. Thereafter, the proportion of
entrepreneur that are selected or—at least—learn to profess that they do it by passion is bound to
grow, just by the selection mechanism.
Obviously, this self-fulfilling property derives from the fact that the keystone rules emerge
out of a genetic algorithm, which amounts to an institutionalization of the field. The process by
which cognitive limitations are resolved by a field calculation (see below) becomes then a force
by itself (iron cage in Weber, 1904) triggering isomorphism (DiMaggio & Powell, 1983 [1991])
across actors—here a relative homogenization of the belief of individual entrepreneurs. This
isomorphism might then be reinterpreted not anymore as an inherently given to human nature,
but more as the powerful social resolution of a hard cognitive problem.
A converging ecology of memes
Similarly, consistent with an institutional perspective, we observe a rich mechanism of
20/35
exchange for entrepreneurial grand theories of action. Entrepreneurs disclose that they did not
elaborate nor test most of their theories of actions. Rather, they heard them from other
entrepreneurs, in a typically contagious manner (Galaskiewicz & Burt, 1991). The archival
exploration demonstrates the common occurrence of entrepreneurs telling the world—hence their
peers—about their grand theories, through the most potent of modern communication means, the
press. Furthermore, their theories are more likely to be diffused if they originate from a
successful entrepreneur—who are often given a greater voice than unsuccessful ones.
Given these characteristics, the individual “theory of action” fits the framework proposed by
Galunic and Weeks (2003) whereby memes—ideas, beliefs, assumptions, values, interpretative
schema, and know-how—can evolve in organizational context similar to genes in biological
systems. Memes therefore spread, vary, recombine and are selected.
The data demonstrates an emergence pattern whereby theories of action have a non-random
distribution, with a concentration of occurrence that is increasing towards a kernel of a very few
rules. This structure is consistent a pyramid view of cognition (e.g. Gavetti & Rivkin,
2007:432), yet it demonstrates that the stratification is not exclusively an individual mechanism,
but rather the result of the individual enacting a set of beliefs in interaction with the environment.
Summary of findings
The keystone rules of entrepreneurs appear as hybrid objects, exhibiting ambidexterity in
multiple dimensions. On an epistemic axis, the rules have the usual limitations of naïve theories
in that practitioners are subjects to inaccuracies and inconsistencies. At the same time, they
exhibit interesting properties by being cognitively dimensioned, generative, polymorphic and
self-fulfilling.
From a structural point of view, the analysis suggests that theories of action in the
21/35
entrepreneurial field are like memes, hosted, selected and varied in an interaction between
individual entrepreneurs and the institutional environment they belong too (here the mainstream
entrepreneurial context of two western countries). The resulting object has the hybrid properties
of being the result of one of the few means to find a locally optimal belief set, while being
simultaneously elaborated in a collective manner, hence having potentially undergone various
alterations due to the social construction processes.
DISCUSSION
Exploring Paradigmatic Structure of Practice through Keystone Rules
This study suggests that keystone rules are a potentially real, even if complex, phenomenon.
On the one hand, as with all practitioners’ beliefs, keystone rules are probably biased or socially
polluted in various ways. Some researchers may argue that, as a result, the most we can establish
are the mechanisms of their social construction. On the other hand, the data and the above
induction suggest that they can have positive properties that would make them adaptive to the
constraints on the rationality of actors. Interestingly, this property matches the description of
science in the lens proposed by Kuhn (1970), except that instead of probing actual scientists, we
have been probing the entrepreneurs as “naïve scientists”.
Kuhn shifted the view of science as a rational and gradual accumulation of valid knowledge
(Popper, 1934 [2002]) to a vision of science as a locus of a social practice, with only limited
changes being conducted because of the existence of a constraining paradigm. Furthermore,
significant changes of practice occur only when the paradigm, a core set of assumptions that
“provide model problems and solutions for a community” (Kuhn, 1970:10), can be altered.
Interestingly, this study suggests a possible similar shift in our thinking about the logic of
action in management studies. Our field might have entertained a Popperian view whereby the
22/35
practitioner should thrive for a goal of perfect rationality informed by the sum of a large body of
scientifically validated rules. Such perspective is advocated most explicitly by an evidence-based
management perspective (Pfeffer & Sutton, 2006) that exhort managers to embrace the scientific
method and results for management practice.
By contrast, consistent with a Kuhnian perspective, we suggest that actors such as
entrepreneurs may be under the spell of a small bundle of beliefs, elaborated and shared
collectively, hence the paradigmatic dynamics suggested by Kuhn apply to these naïve scientists.
By implication, an organizational field such as entrepreneurship could be analyzed as a set of
communities acting according to different paradigms (sets of keystone rules). Since they emerge
by meshing the social construction and rational decision-making, keystone rules—and their
bundling into a paradigm—imply a theoretical hybridization that warrants further research.
Entrepreneurial Paradigms: Effectuations and others
The current study presents great similarities with the effectuation perspective (Sarasvathy,
2001a). First, by focusing on the constraints on elaborating optimal rule sets, we study a key
mechanism towards the elaboration of normative theory at the heart of effectuation paradigm
(Sarasvathy, 2001a). Second, even though the methods do not perfectly match, here qualitative
grounded theory building vs. quasi-experimental lab observations (Sarasvathy, 2001b), they rely
nevertheless on observing practitioners (qualified as “expert” in effectuation research, and here
as experienced) and elaborating on their mental models.
Regarding the substantive findings, effectuation research arouse of a dichotomous contrast
between two paradigms: causation vs. effectuation. Sarasvathy took position against a prevalent
world view based on predicting and planning the future—i.e. causation—and posited that actors
would use the alternative effectuation approach whereby planning is downgraded and attention
23/35
shifts to a effectuation (Sarasvathy, 2001a). The causation logic was described as following a
logic of (Sarasvathy, 2001a:249): start from a goal; determine effective means; take into account
constraints on means; select (usually a maximization of expected returns). The alternate
effectuation logic that emerged can be summarized by its five effectuation principles labeled
idiomatically as follow (Sarasvathy, 2008): “patchwork quilt” (e.g. use existing means);
“affordable loss” (e.g. commit only what one can afford to loose rather than plan expected
returns); “bird-in-hand” (e.g. focus on those willing to commit something to the project);
“lemonade” (e.g. embrace surprises); “pilot-in-the-plane” (e.g. focus on what one can control and
act on).
As far as mapping those principles to the keystones rules emerging here, the overlap is not
perfect. First, the “embrace possibility of failure” keystone rule implies both “lemonade”
(embrace surprises) and “affordable losses”. Second, one can notice that the “self-determined to
the point of irrationality”, with its sub-rules of “not listening to advice” and “being irrational”,
has strong similarities to the key concept in effectuation that not to attempt to plan the future.
Regarding mapping of the other elements, some of the effectuation principles could appear
further down the pyramid of beliefs—typically for instance the focus on existing means was
mentioned in our interviews—but do not appear in our empirics as keystone elements.
Reciprocally, the element of passion (not money) that appears in our keystone rules does not
appear, a priori, to have an equivalent in the effectuation paradigm.
However, effectuation research provides an interesting echo to this study by clearly
structuring epistemology of action around two distinct bodies of beliefs (causation vs.
effectuation). By doing so, it support the need of a distinct construct such paradigm of action as
constituted of bundle of keystone rules. In the view of effectuation scholars, some actors are
24/35
under the spell of a causation paradigm, and some others of an effectuation paradigm. Such
binary structuration of logic action have already been evoked in previous organizational studies,
for instance “theory X vs. theory Y” study that contrasted the world views of managers as a
binary possibility of either “workers intrinsic hate work” vs. “workers find satisfaction in work”
(McGregor, 1957). Along the same line, Miles’ “theories of management” (Miles, 1975; Yoder
et al., 1963) proposed a three-tier model (Traditional vs. Human Relations vs. Human Resources)
and formalized dimensions to qualify those heuristics (assumptions, policies, expectations).
In that sense, effectuation principles may not match clearly the keystone rules that emerge in
our sample but they embody paradigms, and can be clearly contrasted to another paradigm,
causation. Using that framework, we can reinterpret all three paradigms as they could be
decomposed in keystone rules (see Figure 1). For effectuation, we qualitatively group the
principles into bundles (that could be worded as keystone rules). For the paradigm implied by the
keystone rules that emerge here, we simply label it “emergent paradigm”. By doing so, we signal
that the study does aim to a substantive contribution to identify those particular rules, but rather,
aims to contribute by identify the existence of such bundle, the constraint leading their
emergence, and their possible epistemic property. The actual substance at this stage of research
cannot be overplayed nor deserve a particular label.
----- Insert Figure 1 roughly here -----
When comparing the elements of the three paradigms, a pair-wise contrast appears. As
intended originally by Sarasvathy, effectuation and causation can be opposed by the differences
between: “constructing plans” vs. “embracing surprises” and “planning for (affordable) losses”;
“focusing on resources at hand” vs. deriving resources from the goal and the plan that it entails;
“focusing on what one can control and those willing to commit” vs. “optimizing action plans by
25/35
a maximization logic”. Furthermore, effectuation claims agnosticism towards goals vs. the
central role they play in causation paradigm.
In similar manner, the paradigm that emerges from this study can be opposed too to the
causation paradigm: “doing by passion” vs. “maximization logic” (and its hint of monetary
rewards); “embracing failure” and “possibility of irrationality” vs. “the construction of a plan”
and “following an optimization logic”. The notion of resources did not emerge as a keystone
belief in the current paradigm whereby it is a central construct of causation.
Overall, the paradigm that emerges here resembles very much effectuation in its contrast to
the causation paradigm. The notable differences are that effectuation put an emphasis on using
resources at hand (just absent here) and that the current paradigm put a substantive emphasis on
the goal being derived from intrinsic motivations (hinted to be emergent in effectuation). On that
basis, the two approaches land in a very compatible approaches even though substantive
differences remain.
Limitation and Future research
The contrast between those paradigm could constitute interesting direction for future
research. What may appear as differences might indeed, as is hinted here, constitute strongly
similarities, and, in the long run, complementarities among different paradigms. However, the
temptation to add up all rules and assume that they are all valid would contradict the gist of the
current questioning: human actors may have intrinsic limitation making that only one of the
small subsets can be enacted, not only limited by the number of rules but also by the ability to
carry multiple paradigms of intrinsically different spin. For instance, it may be that additional
rule can be carried but that the causation paradigm remains intrinsically incompatible with the
effectuation paradigm, hence difficult to learn, enact, and practice simultaneously for
26/35
practitioners. Future studies could explore the conditions and context that make paradigm the
most relevant model of actors, by contrast to alternative possibility that actors can act on a much
broader base of keystone rules. Entrepreneurial context vs. large organizational context,
education vs., intelligence, stress, etc. may all be factors that could determine the relevance and
breadth of keystone rules.
A limitation of the current qualitative approach was that we attempted only to observe the
existence and character of keystone rules; we did not attempt to measure and quantify the
cognition, since we gave priority to the emergence of grounded theory and the formal reasoning
necessary to interpret it. Future research might explore further ways to measure and validate the
meta-rules beyond just observing their emergence. If establishing validity in absolute sense will
be difficult, one can nevertheless conduct comparative studies of different bundle, both for
descriptive purposes (how the bundle evolve with culture, with experience, etc.) and for
normative purpose (which of bundle A or B is more efficient). These could be conducted in a
positivistic and even somewhat quantitative approach but their nature might require clinical
scholarship in the form of large (quasi)experiments.
Contribution to existing literatures
This study contributes to several management science perspectives. To managerial cognition
literature, the paper contributes by suggesting the importance of keystone rule and their bundling
into paradigms, and by modeling a hierarchy of simple rules (or heuristics). Furthermore, this
study suggests specific characteristics and peculiarities of keystone rules that may differ from
more specific heuristics applied to individual problem (Bingham et al., 2007; Bingham &
Eisenhardt, 2011). This paper adds to institutional theory by revisiting the social construction
interpretation of institution in the instance of shared beliefs in a field such as entrepreneurship. It
27/35
proposes a structuration mechanism (Barley & Tolbert, 1997; Giddens, 1979) linking individuals
to a field construction that does not have to be devoid of function even when it may appear
somewhat dysfunctional (e.g., Abrahamson & Eisenman, 2008; Staw & Epstein, 2000). Finally
this paper also contributes to the greater debate between rigor and relevance in our field (Starkey,
Hatchuel, & Tempest, 2009), with entrepreneurship being particularly subject to the tension
between academic research and practitioner “wisdom” that has often been discounted as limited
and biased.
28/35
APPENDIX A: INTERVIEW PROTOCOL
I. Provide us your Age, Profession, Company and Brief bio. A. … do you have experience in particular as Entrepreneur and/or Investor in Entrepreneurial firms?
II. If A. Investor: "when you choose to invest on an entrepreneur, what characteristics are you interested in?" B. Entrepreneur: "imagine the possibility that young person, a close to you, of becoming entrepreneur. What would you advise that person?" C. In both cases, *after* collecting initial response : "You don't need to provide more than necessary, but I want to make sure I capture your philosophy of entrepreneurship. Anything else?"
III. If the person cited only a limited number, and no generic knowledge (such as "learning everything that can be taught in an MBA"), explore:
A. "when you mention only a few principles, what do you make about the myriads of other things one would actually use?" B. … “e.g. learning entrepreneurship in a formal manner like in books or at school” C. … “or e.g. all the technical knowledge necessary in each business?” 1. If the person minimizes this: "do you assume they are absolutely useless, or just that they will be taken care of ?" D. "Let us go in details about those principles.” List a few of the top ones and ask: “Could you give details why those are important, how they could be crucial to the entrepreneur’s action?" E. “Regarding those rules, how did you establish them?” [try to go one by one] 1. … if not mentioned : a) “Did you read or hear them from someone?” b) Did you test them yourselves?
IV. If morally desirable characteristics were mentioned (e.g. “have passion”, “be honest”): 1. "This characteristic seems morally desirable, such as X, Y. Maybe you do mention even though it might unfortunately actually work the other way?”
V. Look at the master list, and try to identify a few that are not cited. Ask "Consider a few qualities that you did not mention but are sometimes mentioned by other entrepreneurs. What do you make of ..."
29/35
REFERENCES
Abrahamson, E., & Eisenman, M. 2008. Employee-management Techniques: Transient Fads or Trending Fashions? Administrative Science Quarterly, 53(4): 719-744.
Aldrich, H. E. 1999. Organizations Evolving. Thousand Oaks, CA: Sage Publications.
Alessi, G. 1987. Generative strategies and teaching for generalization. Anal Verbal Behav, 5: 15-27.
Argyris, C., & Schön, D. A. 1978. Organizational Learning: A Theory of Action Perspective. Reading, MA: Addison-Wesley.
Barley, S. R., & Tolbert, P. S. 1997. Institutionalization and structuration: Studying the links between action and institution. Organization Studies, 18(1): 93-117.
Bingham, C., Eisenhardt, K., & Furr, N. 2007. What Makes a Process a Capability? Heuristics, Strategy and Effective Capture of Opportunities. Strategic Entrepreneurship Journal, 1(1): 27-47.
Bingham, C. B., & Eisenhardt, K. M. 2011. Rational heuristics: the 'simple rules' that strategists learn from process experience. Strategic Management Journal, 32(13): 1437.
Bruderer, E., & Singh, J. V. 1996. Organizational evolution, learning, and selection: A genetic-algorithm-based model., Academy of Management Journal, Vol. 39: 1322: Academy of Management.
Callon, M., & Muniesa, F. 2005. Economic Markets as Calculative Collective Devices. Organization Studies (01708406), 26(8): 1229.
Chen, X.-P., Yao, X. I. N., & Kotha, S. 2009. Entrepreneur Passion and Preparedness in Business Plan Presentations: A Persuasion Analysis of Venture Capitalists' Funding Decisions. Academy of Management Journal, 52(1): 199-214.
Combs, J. G. 2010. Big samples and small effects: let's not trade relevance and rigor for power. Academy of Management Journal, 53(1): 9-13.
Cyert, R. M., & March, J. G. 1963 [1992]. A Behavioral Theory of the Firm. Malden, MA: Blackwell.
Davis, J. P., Eisenhardt, K. M., & Bingham, C. B. 2009. Optimal Structure, Market Dynamism, and the Strategy of Simple Rules. Administrative Science Quarterly, 54(3): 413-452.
Denrell, J., & March, J. G. 2001. Adaptation as information restriction: The hot stove effect. Organization Science, 12(5): 523-538.
DiMaggio, P. J., & Powell, W. W. 1983 [1991]. The Iron Cage Revisited: Institutional Isomorphism and Collective Rationality in Organizational Fields. In W. W. Powell, & P. J. DiMaggio (Eds.), The new institutionalism in organizational analysis: 63-82. Chicago and London.
Galaskiewicz, J., & Burt, R. S. 1991. Interorganization contagion in corporate philanthropy. Administrative Science Quarterly, 36: 88-105.
Gavetti, G. 2012. Toward a Behavioral Theory of Strategy. Organization Science, 23(1): 267.
Gavetti, G., & Rivkin, J. W. 2007. On the Origin of Strategy: Action and Cognition over Time. Organization Science, 18(3): 20p.
Giddens, A. 1979. Central problems in social theory: action, structure, and contradiction in social analysis. Berkeley: University of California Press.
30/35
Holland, J. H. 1975. Adaptation in natural and artificial systems: an introductory analysis with applications to biology, control, and artificial intelligence. Ann Arbor: University of Michigan Press.
Hongwei, X., & Ruef, M. 2004. The myth of the risk-tolerant Entrepreneur. Strategic Organization, 2(4): 25p.
Kahneman, D., & Tversky, A. 1979. Prospect Theory: An Analysis of Decision under Risk. Econometrica, 47: 263-291.
Kuhn, T. S. 1970. The structure of scientific revolutions, 2d ed.: chp2,3,4,5,ps, 210. Chicago: University of Chicago Press,.
Levinthal, D. A., & March, J. G. 1993. The Myopia of Learning. Strategic Management Journal, 14(8): 95.
MacKenzie, D., & Millo, Y. 2003. Constructing a Market, Performing Theory: The Historical Sociology of a Financial Derivatives Exchange. American Journal of Sociology, 109(1): 107.
McGregor, D. 1957. The human side of enterprise. In J. S. Ott (Ed.), Classic readings in organizational behavior, 1996 ed., Vol. 3: 57-63. Belmont: Wadsworth Pub. Co.
Miles, R. E. 1975. Theories of management: implications for organizational behavior and development. New York,: McGraw-Hill.
Miller, D. 1993. The architecture of simplicity. Academy of Management Review, 18(1): 116.
Ocasio, W. 1997. Towards an Attention-based View of the Firm. Strategic Management Journal, 18(Summer Special Issue): 187-206.
Ocasio, W. 2011. Attention to Attention. Organization Science, 22(5): 1286-1296.
Pfeffer, J., & Sutton, R. I. 2006. Hard facts, dangerous half-truths, and total nonsense : profiting from evidence-based management. Boston, Mass.: Harvard Business School Press.
Popper, K. R. 1934 [2002]. The logic of scientific discovery, Routledge classics.: 27-47, 513. London ; New York: Routledge.
Rivkin, J. W. 2000. Imitation of Complex Strategies. Management Science, 46(6): 21p.
Rivkin, J. W., & Siggelkow, N. 2003. Balancing Search and Stability: Interdependencies Among Elements of Organizational Design. Management Science, 49(3): 290-311.
Sarasvathy, S. D. 2001a. Causation and Effectuation: Toward a Theoretical Shift From Economic Inevitability to Entrepreneurial Contingency. Academy of Management Review, 26(2): 243-263.
Sarasvathy, S. D. 2001b. Effectual Reasoning in Entrepreneurial Decision Making: Existence and Bounds. Academy of Management Proceedings & Membership Directory: D1.
Sarasvathy, S. D. 2008. Effectuation : elements of entrepreneurial expertise. Cheltenham, Glos, UK ; Northampton, MA: Edward Elgar.
Simon, H. A. 1947 [1997]. Administrative behavior: a study of decision-making processes in Administrative Organizations (3rd ed. ed.). New York: Free Press ; London: Collier Macmillan.
Simon, M., & Shrader, R. C. 2012. Entrepreneurial actions and optimistic overconfidence: The role of motivated reasoning in new product introductions. Journal of Business Venturing, 27(3): 291.
Starkey, K., Hatchuel, A., & Tempest, S. 2009. Management Research and the New Logics of Discovery and Engagement. Journal of Management Studies, 46(3): 547.
31/35
Staw, B. M., & Epstein, L. D. 2000. What Bandwagons Bring: Effects of Popular Management Techniques on Corporate Performance, Reputation, and CEO Pay. Administrative Science Quarterly, 45(3): 523-556-556.
Strauss, A. L. 1987. Qualitative analysis for social scientists. Cambridge [Cambridgeshire] ; New York: Cambridge University Press.
Todd, P. M., & Gigerenzer, G. 2003. Bounding rationality to the world. Journal of Economic Psychology, 24(2): 23p.
Tushman, M. L., & O'Reilly, C. A. 1996. Ambidextrous Organizations: managing evolutionary and revolutionary change. California Management Review, 38(4): 8.
Walsh, J. P. 1995. Managerial and Organizational Cognition: Notes from a Trip Down Memory Lane. Organization Science, 6(3): 280.
Weber, M. 1904. The Protestant Ethic and the Spirit of Capitalism (T. Parsons, Trans.). New York, NY: Charles Scribner's Sons.
Weeks, J. R., & Galunic, D. C. 2003. A Theory of the Cultural Evolution of the Firm: The Intra-Organizational Ecology of Memes. Organization Studies (Sage Publications Inc.), 24(8): 1309.
Weick, K. E. 1979. The social psychology of organizing (2d ed.). Reading, Mass.: Addison-Wesley Pub. Co.
Yang, X.-S. 2010. Nature-Inspired Metaheuristic Algorithms: Second Edition: Luniver Press.
Yoder, D., Miles, R. E., McKibbin, L., Boynton, R. E., & England, G. W. 1963. Managers' Theories of Management. Academy of Management Journal, 6(3): 8p.
32/35
TABLES AND FIGURES
TABLE 1
Rules (level 1) by Source
TOTALS: Count % T P 2: BP 3: CP 4: BP 5: JP 6: SP 7: BP 8: CP 9: FP10: P11: P12: P13: P14: P15: P16: P17: P18: P20: P21: P22: P24: P25: TOTALS:3 2 9% T-Attention-Customers 0 0 0 0 0 2 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 37 5 23% T-Attention-People 0 0 0 0 0 0 0 0 0 0 1 1 0 0 1 0 0 1 3 0 0 0 71 1 5% T-Autonomy 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 12 2 9% T-Autonomy-DoListen 1 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2
12 8 36% T-Autonomy-NoListen 0 1 0 0 2 3 0 0 0 0 0 1 1 0 0 0 0 2 1 1 0 0 123 2 9% T-Brave 0 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 1 0 0 0 0 0 32 2 9% T-Complement oneself 1 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 22 1 5% T-Creative/Innovative 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 0 0 0 22 2 9% T-Curiosity 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 23 3 14% T-Decisive 1 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 31 1 5% T-ego aside 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 11 1 5% T-Ethics 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 12 2 9% T-flexible 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 22 1 5% T-Focus 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 0 0 21 1 5% T-Grounded, Make oriented 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 13 3 14% T-Hard Work 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 1 0 0 36 3 14% T-Have low cost base 0 0 0 2 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 65 2 9% T-Idea matter ... or not? 0 0 0 0 0 0 0 0 0 0 0 3 0 0 0 0 0 0 0 0 0 2 53 2 9% T-Intuitive 1 0 0 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 31 1 5% T-Irrationality-Deal with ambiguity, u 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 13 2 9% T-Irrationality-Foolish 0 0 0 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 31 1 5% T-Irrationality-Karma 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 13 3 14% T-Irrationality-Luck 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 1 0 31 1 5% T-Irrationality-No forward 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 11 1 5% T-Irrationality-Serendipity 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 11 1 5% T-Know Domain 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 13 3 14% T-Know your limit, ask for help 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 1 0 0 33 2 9% T-Leadership 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 1 0 0 3
16 8 36% T-Learning not from school 0 0 0 1 0 0 0 2 2 1 0 0 0 1 0 0 5 2 0 0 0 2 161 1 5% T-Limitations 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 11 1 5% T-Long Term 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 11 1 5% T-Manager-Not 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 15 4 18% T-Manager-OK 0 0 0 0 0 0 0 0 0 2 0 0 0 1 1 0 0 0 0 0 0 1 5
10 7 32% T-Mistake-Learn from 0 0 0 0 0 0 0 0 0 0 1 1 1 1 0 0 3 1 2 0 0 0 1012 11 50% T-Mistake-Relentness 1 0 0 1 1 0 1 0 0 0 1 1 1 1 0 0 0 1 2 0 1 0 12
8 7 32% T-Mistake-Risk Takers78521 1 0 0 0 2 1 0 0 0 0 0 1 0 0 0 0 0 1 1 0 1 0 816 12 55% T-Mistakes-do them 2 0 0 1 2 2 0 1 1 0 2 1 0 1 0 0 1 0 1 0 1 0 16
2 2 9% T-Mistakes-Risk-No 0 0 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 0 0 0 0 0 25 4 18% T-Motivation-Money 0 0 0 0 1 0 0 0 0 0 0 0 0 1 0 0 0 1 2 0 0 0 50 0 0% T-Motivation-Passion-Non 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
16 10 45% T-Motivation-Passion-Yes 0 0 0 3 0 0 1 0 0 0 1 2 0 0 0 1 2 1 2 1 2 0 161 1 5% T-no Hobby 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 11 1 5% T-No need for status 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 11 1 5% T-Not for the faint hearted 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 13 2 9% T-not person, can be taught 0 0 0 0 0 0 0 1 0 2 0 0 0 0 0 0 0 0 0 0 0 0 31 1 5% T-Open-minded 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 15 5 23% T-Personality type 0 0 0 0 1 0 0 1 0 0 0 1 0 0 0 0 0 0 0 1 0 1 52 2 9% T-purpose 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 21 1 5% T-Questions authority 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 13 2 9% T-Resistant/Erergy 0 0 0 0 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 31 1 5% T-Resources 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 11 1 5% T-Search Opportunities 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 11 1 5% T-Simple idea 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 11 1 5% T-Smart 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 11 1 5% T-Start Early? 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 11 1 5% T-Thinking long term 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 11 1 5% T-Value 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 12 2 9% T-Vision 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 2
198 22 100% TOTALS: 10 1 3 16 19 9 2 5 3 5 11 15 3 10 2 3 16 14 21 7 11 12 198
33/35
TABLE 2
Grouping of Heuristics
Grouping Count Heuristics TD-Mistakes 56 [T-Mistake-do them] [T-Mistake-Learn from] [T-Mistake-
Relentness] [T-Mistake-Risk-No] [T-Mistake-Risk-Takers] TD-Motivations (Passion vs. money)
27 [T-Motivation] [T-Motivation-Money] [T-Motivation-Passion-Non] [T-Motivation-Passion-Yes]
TD-Autonomy 21 [T-Autonomy-DoListen] [T-Autonomy-NoListen] TD-Not from school
21 [T-Autonomy-DoListen] [T-Autonomy-NoListen]
TD-Attention 14 [T-Attention-Customers] [T-Attention-People] TD-Irrationality 14 [T-Irrationality-Deal with ambiguity, uncertainty] [T-
Irrationality-Foolish] [T-Irrationality-Karma] [T-Irrationality-Luck] [T-Irrationality-No forward] [T-Irrationality-Serendipity]
TD-Strengh 8 [T-Brave] [T-Not for the faint hearted] [T-Resistant/Erergy]
TABLE 3
Meta-Heuristics (level 2) by Source
Count % P 2 P 3 P 4 P 5 P 6 P 7 P 8 P 9 P10P11P12P13P14P15P16P17P18P19: P20: P21: P22: P23: P24: P25: TOTALS:
16 70% *TD-Mistakes 4 0 0 1 5 3 1 3 1 1 0 4 3 2 3 0 0 4 3 6 0 3 0 0 4712 52% *TD-Motivations (passion v 0 0 0 3 1 0 1 0 0 0 0 1 2 0 1 0 1 2 2 4 1 2 0 0 2111 48% *TD-Autonomy 1 1 1 0 2 3 0 1 0 0 0 0 1 1 0 0 0 0 2 1 1 0 0 0 157 30% *TD-Attention 0 0 0 0 0 2 0 0 0 0 0 1 1 0 1 1 0 0 1 3 0 0 0 0 108 35% *TD-Learning not from sch 0 0 0 1 0 0 0 0 2 2 1 0 0 0 1 0 0 5 2 0 0 0 2 0 166 26% *TD-Irrationality 0 0 0 5 1 0 0 3 0 0 0 1 0 0 0 0 0 1 0 0 0 2 0 0 135 22% *TD-Strengh 0 0 0 0 3 0 0 1 0 0 0 0 2 0 0 0 0 1 0 0 0 0 1 0 8
23 100% TOTALS: 5 1 1 10 12 8 2 8 3 3 1 7 9 3 6 1 1 13 10 14 2 7 3 0 130
15 65% autonomy+irrationality 1 1 1 5 3 3 0 4 0 0 0 1 1 1 0 0 0 1 2 1 1 2 0 0 28
34/35
TABLE 4
Description of Keystone Rules
Meta-heuristic (level 2) Heuristic (level 1) Example citations
Be prepared to meet failure “One of the most important attributes to being a succ(essful entrepreneur is the ability to learn how to be wrong and fail fast” (Th)
Learn from your failure
“Make the mistakes and learn from them. So what if your business fails? Your next one will be even better than your first, and the next one after that will be even better again. Essentially, treat it as a learning experience, one that teaches lessons money could never buy. And over time wisdom shall enable more fruitful outcomes” (Gu)
Embrace possibility of failure
Be able to reboot after failure “The trait of not quitting, ever. You can never give up as an entrepreneur. If their first business goes under, they start another one. If their seventh business goes under, they start another one, and so on.” [Bro)
Do it by passion (not money)
“In my view wanting to be an entrepreneur just for the purpose of getting rich is not a strong enough motive to survive the journey ahead” (Kha)
Do not listen to advice “The difference between founders and professional managers is that founders are stubborn about the vision of the business” (Bez)
Be self-determined to the point of irrationality
Be Irrational “the best [entrepreneurs] are a bit contrarian, sometimes unreasonable” (Bot)
35/35
FIGURE 1
Comparison of three exemplar paradigms