It's Time to Rethink Culture

Venereal disease in the Middle Ages and early Renaissance was thought by the scientists of the day to be caused by the sin of fornication, seen as just punishment for sexual transgression. Such a diagnostic frame could have only been possible in an age where religion explained most natural phenomena. In similar fashion, the recent ethical lapses at General Motors leading to the recall of 2.5 million cars has been likened to a failure of GM’s corporate culture. The failure of Enron in the early 2000s, and many other firms is similarly attributed.

Corporate culture today is the summary explanation for bad behavior, ethical lapse, poor performance, low morale, failed strategy, and many other problems, as well as the source of enduring success and un-replicable competitive advantage. As put recently by The Economist: ‘Culture’ is the mot du jour in the business world.

Perhaps this is why most CEOs and management consultants believe culture "shaping" is a core part of the job description. Embedded in this line of thinking is an assumption: culture can be engineered, managed and molded for competitive advantage, like equipment, financial resources, brand, or strategy. Once you get your culture "right", success will follow. This assumption is so deeply ingrained in mainstream management thinking that to suggest it might be, at best, a theory without a lot of supporting evidence, or at worst, deeply wishful thinking, is tantamount to heresy.

Before you commit me to the burning stake, let me offer a few reasons why.

The Problem

Organizational norms and values were referenced as early as the 1950s (see Homans, 1950, or Selznick, 1957). But organizational culture per se as a means of managing and controlling organizations is a much more recent management concept traceable to the 1970s (see, for example, Kunda, 1992). If one considers the cultural and cognitive anthropology literature, arguably the field most concerned with the study of culture, there is very little in it about culture “shaping”.

Beyond the confines of what I call the culture “industry” – the management consulting firms and business schools who offer the thinnest of case-based or reductive quantitative evidence of culture change -- there is little academic consensus on what culture is or whether it, in fact, can be managed. Despite nearly 40 years of research, academics still disagree on even the most fundamental questions: What is culture? Can it be measured, and if so, on what basis? These questions are so contested in the literature they are characterized by Joanne Martin and Peter Frost (2011) as “culture wars”, leading The Economist (2014) to conclude (one of the few in the mainstream press to do so) that the term is so vague that “culture based explanations are often meaningless” (January 11, p. 72).

Not surprisingly, the few serious academic studies on culture change suggest ambiguous results, anecdotal evidence, or no major effects on anything other than espoused values. The problem is not the concept; culture is a real phenomenon, and all organizations have them. The problem is the way culture is defined, measured and operationalized. The lack of consensus on these items leads to a kind of Wild West of culture practice where nearly anything goes, such as when the CEO sends a memo to the company saying what the culture is going to be. Not surprisingly, most attempts at culture change don’t yield desired results and waste inordinate amounts of time, money and resources. Yet, for reasons I outline below, many leaders and consultants think this is a natural and even essential part of the CEO agenda.

In following paragraphs I will explain why it is time to rethink this approach. This piece briefly examines several assumptions characterizing mainstream culture research and practice, and provides a brief outline of an alternative “cognitive” paradigm that advances culture beyond these constraints. This paradigm offers a compelling alternative to enable more effective and sustainable culture interventions.

What’s Wrong with Culture?

The idea that culture is a source of competitive advantage is built on a set of problematic assumptions that have received little critical examination.

The assumption of “thingness’

We tend to perceive social categories such as culture existing naturally in the world. For example, M&A is often characterized by metaphors such as FIGHTING, MATING or EVOLUTIONARY STRUGGLE. These metaphors are embedded in multiple narratives and tend to make “real” the sense mergers really are a struggle to evolve (etc.). This same tendency, as the social historian William Sewell Jr. puts it, when applied to culture “ontologizes” it, rendering it as something real. This phenomenon is aided by a major feature of human cognition, analogical reasoning, which enables thinking about abstract things by analogy to physical things. Our language reflects this through the many metaphors we use to talk about culture, metaphors that over time become so conventionalized and routine we take abstractions like cultures to be physical things. Nowhere is this seen as clearly as in the Corner Office column in the New York Times that weekly publishes interviews with corporate leaders on how they “view their cultures” and what they are trying to achieve “with” their cultures”. The very posing of these kinds of questions, questions so routine the underlying assumption of culture as a real, physical thing is obscured from all but the most attuned observer, reinforces the notion of culture as something to be physically manipulated. Once so conceptualized, culture shaping is easily conceived of as reasonable, even logical.

The assumption of stability

When culture is assumed to be a physical thing, we ascribe physical properties to it. One of these is stability. That is, cultures are simply ‘out there’, lying in wait to be examined and measured and manipulated.

Management science and mainstream culture practice makes the assumption the social world is naturally and necessarily made up of discrete quantifiable units. Without assumed stability, quantifiable measurement is impossible. With assumed stability, causality can be “seen” in a statistical distribution of “cultural” traits presumed as causal explanations of behavior. This is an assumption of most quantitative studies of culture.

Quantitative reasoning in the social realm however is based on language as well as symbolic and representational practices, many of which are generated by underlying paradigms which are socially constructed to begin with (e.g. culture causes certain behaviors). What is measured, by the very fact of its measurement becomes thought of as real and valid. What is seen as correlations in data thus are taken as causal because the basis of the measurements are assumed to be stable. As actors adapt to these patterns this contributes to the idea that culture is a real and objective thing with stable properties and features.

At issue is not quantitative reasoning. At issue is the problematic assumption of culture as a thing, and the inference of stability that follows from it. As anthropologists like to say, cultures do not stand still for their portraits.

The assumption culture is reducible

With the inference of stability comes the tendency to reduce culture to one or two variables. The most common of these are personality (writ large) or values.

Culture as personality

Freud believed cultures could be analyzed. The Culture and Personality school in anthropology saw personality and culture as analogous. But some organizational researchers see the relationship in literal terms, proposing, for example, that over time organizations become relatively homogenous based on the personalities of the people in them, thereby accounting for culture. Or similarly, that the personality traits of the founder or the CEO, such as the need for achievement, become manifest within the strategy and structure (and presumably the culture) their organizations adopt.

To assume cultures are homogenous melting pots of personality is to assume cultures are monoliths and all people within them over time become like each other. On the face of it that seems just plain wrong; the direction of causality, for one, is bi-directional. People influence culture as much as culture influences people.

And traits such as achievement are so generalized and well distributed in the CEO population that attribution of them to culture is a tautology. Most CEOs, by their own admission, are achievement oriented; they would not be CEOs otherwise. Even in start-ups where one might expect to see a direct relationship between founder personality and cultural attributes, studies have found no clear relationship, especially as organizations grow (see Siehl, 1985).

To conflate culture with personality also requires a leap of faith. The faith is that personality traits (e.g. agreeableness, extroversion, conscientiousness) reflect underlying psychological realities. Cognitive anthropologists such as D’Andrade (1995), for example, have shown that terms describing personality traits in questionnaires can be interpreted on the basis of similarity in meaning (e.g. friendly positively correlates with sociable; cleverness with inventiveness). What is contested is not the existence of personality traits, but that tests that link traits with culture may be measuring the same thing, and therefore not really correlating anything at all.

Culture as values

Culture, of course, has something to do with values. But the question is can it be reduced to values?

If values are to function as normative “shoulds” the values need to be internalized by everyone in the organization. But anthropologists have shown that values and norms tend not be internalized unless they are already well socialized. This suggests the larger and more heterogeneous an organization is in terms of dominant professional orientation, ethnic composition, or regions of the world in which it operates (etc.), the likelihood of values being universally shared and socialized will be quite low.

Values also may be so accepted and routine within a cultural community they may be enacted without awareness. For example, the Baktaman people of New Guinea have strong norms of sharing that lead to equal living standards. But they do not have an espoused value commending generosity. The Balinese, on the other hand, have large differences in wealth but have an explicitly stated value of generosity (Strauss, 2012). Characterizing culture based on the values a community ostensibly holds is to make the flawed assumption that culture is based on accepted norms, whereas the values espoused may be reactions or compensations for deeper, more pervasive and perhaps contradictory structures. To base culture on what is espoused may well be to mistake the trees for the forest.

Another issue is ethical: whose values? The CEOs? The executive team’s? The consultant who creates a survey purporting to measure values based on his or her own pre-defined and context-free determination of what the values are to begin with? Modern global organizations are not comprised solely of North American or European Caucasians. So when a North American or European company asserts its values, what message are they sending its global workforce?

And values are surprisingly difficult to measure. First, the concept is polysemous. It can mean goodness, preference, price, worth, or moral right. In some languages like Japanese or Vietnamese, there are no adequate translations of the term.

In addition, anthropologists and cultural psychologists have known for years that attribution and categorization systems vary across cultures. For example, when discrete features of cognition are considered such as systems for color, plant classification, kinship, or even conceptions of the self, considerable cultural variation is seen.

Another issue is ideation: value ratings measure idealized notions about how the world should be, rather than necessarily how it is actually perceived. Differences in values rest on differences on what one is expected to hold as a member of a society or institution rather than personal and subjective responses. Scales such as individual-collective or altruism-self interest are measures of perceived ways one should live. And responses to value surveys tend to fail to discriminate sharply among values that are perceived to be similar to each other.

The assumption of boundaries

Whose culture are you shaping when you profess to do so?

Consider the case of a Bengali software engineer living and working in Palo Alto. This person participates simultaneously in American culture when going to the supermarket or dealing with the cable company or commuting to work, but also in Bengali culture when with his family or raising his children, and with a software engineering culture at work. Each of these cultures has its own language, symbols, judgments, authority figures, norms and underlying paradigms, and the physical, social or political boundaries of each are blurred and overlapping, such as when our engineer is at home with his wife watching TV while working on his laptop on a project for work. The question is when he receives a culture survey with a battery of items linked to pre-defined cultural dimensions (as he is likely to receive at some point if he works in large global corporation), which of the many cultures in which he participates will this instrument actually be measuring? And how would those administering such a survey know?

People belong to multiple cultures and communities, so it is difficult to presuppose cultural boundaries. Not surprisingly, studies that treat organizational culture as a monolith comprised of one or two variables obliterate the concept of boundaries altogether. Such generalizations about culture ignore differences of race, class, gender, or professional orientation. ‘Culture’ does not end once you step across national borders or off corporate property. Systems of belief are not unchanging - they are in flux as actors draw on different resources and repertoires of knowledge to make sense of and function within their environments.

The assumption causality

Lastly, mainstream culture practice makes the assumption culture is causal in fostering desired behavior. But demonstrating a discrete causal link between organizational culture, particularly values, and specific performance outcomes is exceedingly difficult, particularly because determining which are the independent and dependent variables and the direction of causality is very hard.

Take but one well-publicized case, that of Southwest Airlines. Culture-shapers have argued its Southwest’s values of teamwork, family, individual initiative, fun (etc.) that render it so successful. What is missed in this line of reasoning is that such values can be easily ascribed to Southwest’s core purpose of being an “airline for the common man” (interview with Herb Kelleher, 2008, in Heskett, 2011). This unique and differentiated core purpose led Southwest to the myriad of strategic choices necessary to be a successful low cost carrier, such as a standardized fleet, fast aircraft turnaround times, simple and innovative pricing, and so forth. It could be argued values such as teamwork and initiative (etc.) came about as a result of this core purpose and were actually posited to enable it. The point is arguments for cultural causality always come back to such circularity. Values may be a result of underlying factors or instituted as idealizations to enable something more fundamental such as core purpose or mission. In assuming values (or other discrete variables) to be causal in performance rather than an outcome of antecedent choices, culture shapers may be dealing with symptoms, not root cause.

The Result

The culture industry for the most part ignores these concerns. Why? For two seductive reasons that are hard to dispel.

The seduction of normative control

CEOs, of course, want to drive financial performance as well as leave a legacy of positive change. Culture is believed to be simply one of the tools through which to do so. The zeal for using culture as a magic bullet for normative organizational control and competitive advantage results in the uncritical acceptance of culture as an ontologized, stabilized, reduced, quantified, personified, idealized and causal instrument for change.

The seduction of self-enhancement

The “self enhancement bias” (Martin, 2002) is the belief leaders have an inordinate and long lasting impact on their organization’s culture based on their own values, ideologies or personality preferences. Again, this leads to reduction and oversimplification. Most leaders who believe they can causally shape culture in any system more complex than a small start-up would be on the wrong side of the anthropological research.

Consequences

By greatly oversimplifying and reducing culture, it risks becoming little more than corporate sloganeering in the form of tag lines and posters, or pool tables and Friday beer parties to make the workplace “fun”. These efforts may help with employee engagement, but have little impact beyond surface effects. And the concept of a distinct culture yielding competitive advantage because culture is believed to increase resource heterogeneity, making it hard to duplicate by competitors is also suspect: companies in the same industry tend to be more alike than companies in diverse ones, rendering such effects muted at best.

Leaders have the correct instinct to attempt to intervene in their own cultures. But most are ill equipped to do so, because culture is the least understood and most complex domain through which to enact change. Which is why we are where we are today: culture as a concept has been rendered almost meaningless, and many culture change interventions have little to show for their efforts. If organizational culture change continues to fail to deliver a return on investment, leaders will eventually abandon the concept altogether.

Meanwhile, the opportunity to improve organizations and, yes, business results in a sustained and predictable way through culture continues to lie in wait. Ironically, much of the power of culture as a resource is lost when it is reduced and simplified. Which is why, far from abandoning the concept, we need better theory, as well as better tools and more sophisticated and systemic interventions.

Rethinking Culture

The last 30 years in cognitive anthropology, linguistics, psychology, and neuroscience have yielded an abundance of theory and research to offer an alternative paradigm for culture, and as a result, new intervention possibilities. Space precludes but a brief introduction.

This new paradigm begins with the idea cultural meaning is largely a cognitive phenomenon, cognitive not only in the sense of knowledge but also, and perhaps largely based on shared tacit knowledge. Culture is loosely but distinctively ecologically determined, underwritten by cognition grounded in the functional, technological and social forces inherent in work, and the production of systems of meaning that accompany work.

Put another way, the work professionals in organizations do, their basic orientation to problems, the axioms they invent, the tasks they determine as critical, the processes and routines they enact, the resources they marshal, the tools and technologies and artifacts and social conventions (think: values) they construct or appropriate - all of these – influence the creation and propagation of organizational culture.

Culture as shared mental models

What makes this possible is sensemaking: we make sense of our worlds by drawing -- most often unconsciously - from shared mental or “cultural” models that have been deeply shaped over time by our physical and social environment (Strauss & Quinn, 1997).

As Harvard's Gerald Zaltman puts it, “nearly all cognition is below our level of awareness”. A cultural model is a set of shared and mostly tacit assumptions about how the world works. These assumptions underwrite much of what we collectively believe to be true, how we make sense of things, how we act, what we value, and how we speak (think: framing). They underpin much of what happens “culturally” inside an organization – how problems are conceptualized, how assumptions are made, what gets privileged as knowledge, how organizations are structured, what gets represented to the world (e.g. branding), what defines success, and even how power is conferred – in short, in most of the ways humans think and behave in collective enterprise. Cultural models are a kind of operating system for culture, permeating nearly all facets of corporate behavior by underwriting the full range of cultural experience, meaning and expression.

In other words, organizational culture is a collection of cultural models that members use, most often unconsciously, to make sense of their world.

Implications

At this point you might ask, ‘so what?’ Culture as mental models sounds good, but how is this any more helpful than thinking of culture as values or personality, or any other single construct?

Culture as mental models can revolutionize culture practice. When culture is recast as a systemic cognitive phenomenon rather than reduced to one or two variables, the focus expands from “how we do things around here” to how we determine what those things are to begin with. The focus shifts from normative values or behaviors to interventions targeted at the underlying and taken-for-granted assumptions that underwrite all cultural action, such as how we formulate problems, define success, make sense of ambiguous data, grant privilege, formulate strategy, shape processes and routines, or decide what is or is not important and so on. These kinds of interventions are at a level of cognition beneath that of values or norms (etc.) and therefore are more pervasive, meaningful, and long lasting.

Ed Schein proposed in 1986 that culture is a system of basic assumptions. Cognitive science is proving him right. Organizations do things largely based on the cognitive predispositions of its members, predispositions which are largely shaped by their professional /occupational orientation and by the strategic tasks and choices necessitated by the furtherance of the enterprise itself, especially in its differentiated and unique purpose (to the extent it has one. Those that do tend to have very “strong” cultures). But these predispositions run in the background, largely invisible to actors, and are not eligible to be easily “shaped”. They delimit and constrain choice, rendering organizations less the product of rational action and more the product of collective sensemaking that continually but (mostly) unconsciously draws upon these resources to explain ‘what is’ and to orient action.

This is why manufacturing companies have such a hard time turning themselves into software companies, or why hospitals struggle to operate as entrepreneurial start ups, or why investment banks cannot get away from information arbitrage as a fundamental strategy even when they strongly espouse “main street” ethics, or why conglomerates have a hard time after divestiture turning themselves into growth companies. It doesn’t mean cultural change is impossible. It means it is predisposed to occur according to the cognitive ecology of its members in complex interaction with the broader macro environment. It is only when these shared mental models are brought into view that their pervasive effects can be seen and sustainable change begin to take place.

And therein lies a new role for leaders who want to change culture. The new role is not to imprint personal ideology but to challenge the basic assumptions on which the organization operates. This is a big shift and requires fundamentally new skills – skills as change agents and provocateurs that use role power to hold the mirror up to their organizations and question the most basic assumptions on which the firm operates. This is, of course, not easy, especially when an organization has had some success. But by bringing to light and, as necessary, changing the many tacit, taken-for-granted assumptions that underwrite how the enterprise "thinks" and fundamentally makes sense of its business, in all its manifestations, sustainable culture change can take place.

Suffice to say, we are in the early days of learning about the “cultural mind”. The implications for organizations are truly profound.

(The preceding is excerpted from a forthcoming book)

Further Reading

A short bibliography of some of the research referenced in this article is here:

1. Alvesson, M., & Sveningsson, S. (2008). Changing organizational culture: Cultural change work in progress. Abingdon: Routledge
2. Bruner, R. (2001). Does M&A pay? A survey of evidence for the decision-maker. Retrieved from: https://scholar.google.com/scholar?hl=en&q=M%26A&btnG=&as_sdt=1%2C5&as_sdtp
3. Cartwright, S., & Cooper, C. L. (1993). The role of culture compatibility in successful organizational marriage. The Academy of Management Executive, 7(2), 57-70.
4. D'Andrade, R. (1995). The development of cognitive anthropology. Cambridge: Cambridge University Press
5. Gertler, M. (2004). Manufacturing culture: The institutional geography of industrial practice. Oxford: Oxford University Press
6. Kelly, J., Cook, C., & Spitzer, D. (1999). Mergers and acquisitions: A Global Research Report (Unlocking Shareholder Value: The Keys to Success). A Study by KPMG.
7. Kunda, G. (1992). Engineering culture: Control and communication in a high-tech corporation. Philadelphia: Temple University Press.
8. Larsson, R., & Finkelstein, S. (1999). Integrating strategic, organizational, and human resource perspectives on mergers and acquisitions: A case survey of synergy realization. Organization Science, 10(1), 1-26.
9. Leonardi, P. & Jackson, M. (2009). Technological grounding: Enrolling technology as a discursive resource to justify cultural change in organizations. Science, Technology and Human Values, 34, 393-418. Doi: 10.1177/016224390832877
10. Hesketh, J. (2011). The culture cycle: How to shape the unseen force that transforms performance. Upper Saddle River, NJ: FT Press
11. Homans, G. (1950). The human group. New York: Harcourt, Brace, & World)
12. Hutchins, E. (2005). Material anchors for conceptual blends. Journal of Pragmatics, 37, 1555-1577. Retrieved from https://scholar.google.com/
13. Hutchins, E. (1995). Cognition in the wild. Cambridge: MIT Press
14. Maltseva, K. & D'Andrade, R. (2011). Multi-item scales and cognitive ethnography. In: D. Kronenfeld, G. Bennardo, V. de Munck, and M. Fischer (Eds.). A Companion to cognitive anthropology. (pp.153-170). Malden, MA: Wiley-Blackwell.
15. Martin, J. (2002). Organizational culture: Mapping the terrain. Thousand Oaks: Sage.
16. Martin, J., Frost, P. (2011). The organizational culture war games: a struggle for intellectual dominance. In M. Godwin and J. Hoffer Gittell (Eds.), Sociology of organizations: Structures and relationships (pp. 559-621). Thousand Oaks: Sage. Retrieved from: https://scholar.google.com/
17. Miller, D., & Dr?ge, C. (1986). Psychological and traditional determinants of structure. Administrative science quarterly, 539-560.
18. Mottola, G. R., Bachman, B. A., Gaertner, S. L., & Dovidio, J. F. (1997). How Groups Merge: The Effects of Merger Integration Patterns on Anticipated Commitment to the Merged Organization. Journal of Applied Social Psychology, 27(15), 1335-1358.
19. Ogbonna, E., & Wilkinson, B. (2003). The false promise of organizational culture change: A case study of middle managers in grocery retailing. Journal of Management Studies, 40(5), 1151-1178.
20. Selznick, P. (1957). Leadership in administration: A sociological interpretation. Evanston, IL: Peterson & Co.
21. Sewell, W. H. (2005). Logics of History: Social Theory and Social Transformation. Chicago: University of Chicago Press.
22. Siehl, C. (1985). After the Founder: An opportunity to manage culture. In P. Frost, L. Moore, M.R. Louis, C. Lundberg, & J. Martin, (Eds.). Organizational culture. (pp. 125-140). Newbury Park: Sage Publications.
23. Schein, E. (1992). Organizational culture and leadership (2nd Edition). San Francisco: Jossey-Bass.
24. Schneider, B., & Smith, D. B. (2004). Personality and organizational culture. Personality and Organizations, 347-370.
25. Strauss, C., (2012). Making sense of public opinion: American discourses about immigration and social programs. New York: Cambridge University Press.
26. Strauss, C. & Quinn, N. (1997). A Cognitive Theory of Cultural Meaning. Cambridge: Cambridge University Press.
27. Swidler, A.(2001). What anchors cultural practices. In T. Schatzki, K. Knorr-Cetina, and E. Von Savigny (Eds.). The Practice turn in contemporary theory. (pp. 74-92). London: Routledge.
28. Triandis, H. (1996). The psychological measurement of cultural syndromes. American Psychologist. Vol. 51. No. 4. pp.: 407-415.