Understanding Cultural Difference Management through Charles Taylor's Philosophy: Case Studies from the Food Processing Industry

† Signature indicate personal opinion and do not represent official positions of the department of Foreign Affairs, Trade and Development Canada or the Government of Canada. Abstract: In this paper, we use the work of the philosopher, Charles Taylor, to investigate the role of culture on internationalization decisions. Using parameters related to key constructs such as positive liberty, social ontology, expressivism, civic republicanism and common spaces, we look at how culture influences the decisions regarding corporate international expansion. This framework was applied in a multi-interview design in four firms from the food processing industry from France and Canada. Results showed an obvious sensitivity to cultural difference and that managerial practices surrounding this issue tended to be intuitive and emergent. These practices were not crystallized in the form of a conscious and deliberate organizational strategy for dealing with cultural difference when planning foreign market entry. Our findings triggered further reflections on managerial implications such as the importance of searching more explicitly for cultural and organizational anchors when reviewing location factors.


Introduction
Because of growing internationalization of corporate activities, individuals from different cultures are in more frequent contact than in the past.Scholarly interest has therefore been stimulated in studying the concept of culture and its impact on the formulation of corporate strategies.Of course, management researchers have long been interested in the notion of culture, but their research has been applied mainly to detecting whether or not culture is a factor (the if) in influencing business decisions.Relatively few have tried to open the black box of decision-making to see how the cultural variable influences the formulation of decisions regarding corporate penetration of foreign markets.In our view, to go beyond the if and better reveal the how, we have to look past the relationship between executives' attitudes and their decisions.We must drill down to the level of the values that lie underneath the decision maker's attitudes toward culture.This suggests further opening research to contributions from disciplines other than management.We test this intuition by constructing a grid inspired by the work of philosopher Charles Taylor.We then apply it to the internationalization strategies of four firms in the food processing industry.Our results show an obvious sensitivity of executives for the issue of cultural difference (the if), but also show that managerial practices surrounding that issue tend to be an intuitive and emergent process (the how) and are not crystallized in the form of a conscious and deliberate organizational strategy for dealing with cultural difference when planning foreign market entry.We conclude nonetheless, though cautiously, that management scholarship can gain from opening itself up to contributions from other fields.

Two Cultural Paradigms in International Business Studies
Definitions and conceptualizations of culture are countless.Culture has been defined at the organizational level, where as corporate culture it refers to the beliefs and values of top-level managers [1].At the national level, Hofstede [2] defined national culture as a collective programming of the mind.Several scholars have deplored that corporate and national cultures "seem to exist in a state of splendid isolation" from each other [3].Adler and Jelinek were among the first to link both levels by defining "culture, whether organizational or national, as a set of taken-for-granted assumptions, expectations or rules for being in the world" [4].We make that definition ours.
Research on the role of culture in international management is extensive.From a theoretical point of view, the field is divided fundamentally between two major perspectives: the functionalist and the interpretive paradigms [5,6].The functionalist perspective proposes an objective vision of the organization.It focuses on establishing causal relationships between culture and organizational decisions.The interpretive perspective underlines the subjectivity of individuals and organizations' capacity or adaptation.

The Functionalist Paradigm
The functionalist paradigm covers the majority of the research on the influence of the culture of the country of origin or host country on corporate activities.With his book Culture's consequences [2], based on a vast survey of attitudes, Geert Hofstede laid a foundational cornerstone.He placed psychology in the forefront of disciplines that have influenced the functionalist paradigm for a large body of research.
Several studies addressed strategic decision-making from the angle of choosing how to enter a market.Kogut and Singh [7] examined the relationship between levels of cultural distance, assessed on a composite index based on Hofstede's five dimensions, and different entry modes.Barr and Glynn [8] studied the interpretation of threats and opportunities according to strategists' national culture.However, Tihanyi, Griffith, and Russell [9] came to the conclusion that there is no obvious link between cultural distance and choice of mode of entry.Among numerous others, Hutzschenreuter and Voll [10] saw the distance between the firm's and the host country's national cultures as an additional layer of complexity influencing the performance of internationalization strategies.Culture variables were also suggested as key factors of the performance of cooperative ventures [11][12][13][14].Shenkar [15] has criticized these studies for concentrating too much on cultural distance rather than seeking to understand what brings cultures together.
What is striking is the variety of concerns of the authors who have made culture a central area of research.Among others, culture was found to influence international labour [16,17], the efficiency of negotiations [18], the emergence of cooperation [19], the performance of acquisitions [20,21], the compatibility of executives in firms [22], the influence of culture on adherence to the strategic orientations [23,24] or how executives interpreted threats from the business environment [25].Mosakowski and Earley [26] proposed that management styles be adapted to individual's perception of time, which is influenced by their culture.Newman and Nollen [27] suggested that an organization's performance is a function of the harmony between the national culture of the country where it is based and its management practices.Nohria and Ghoshal [28] suggested that a multinational's different divisions should generate their respective core cultural values rather than following those of the head office.Ralston, Holt, Terpstra and Kai-Cheng [29] considered whether it was preferable or not for a multinational to impose a single management style on all of its entities.They argued for a level of crossvergence which allows foreign subsidiaries to develop according to their respective cultural logics.Peterson and Smith [30] emphasized the connection between national culture and appropriate management practices, a link which Shaw [31] had already highlighted in his study on relations between executives and subordinates separated by their respective national cultures.Waldman et al. [32] showed that many of Hofstede's dimensions proposed predicted the perception that executives will have of their social responsibilities.
Also following in Hofstede's footprints, Lenartowicz and Johnson [33], as well as Leung et al. [34], identified different groups of Latin American countries that share a similar cultural horizon.Boyacigiller and Adler [35], Kiggundu, Jorgensen and Hafsi [36], Hafsi and Farashahi [37] and Peterson et al. [38] were interested in the systematic application of Western management theories in developing countries.All of these researchers rejected the thesis of the inevitable cultural homogenization of management practices.Another group of authors called for an ethnographic contribution to management research.Appadurai [39] proposed the concept of transnational anthropology, whereas Hannerz [40], Inda [41], and Marcus [42] stressed the usefulness of multi-site and multinational observation in organizational ethnography.Some authors have been interested in the issues involved in transferring human or material resources into foreign cultural milieus.In comparing two types of intercultural training in preparation for foreign mandates, Earley [43] concluded that it is effective to give candidates a reading list and offer intercultural encounters before they depart, in order to facilitate their integration into their new workplace.Griffith, Hu, and Ryans [44] investigated whether the variable of cultural similarity facilitated the standardization of processes in relations between individuals of the same culture and of different cultures.Kedia and Bhagat [45] and Steensma et al. [46] used dimensions inspired by Hofstede to develop conceptual frameworks in an attempt to explain the greater or lesser effectiveness of technological transfers.
Within this functionalist paradigm, some questioned the hegemony of Hofstede's methodology.McSweeney [47] challenged the realism of a causal analysis for understanding a phenomenon as rich and complex as national culture.Tung [48] underlined two important limitations of studies devoted to the values of workers in different countries: the premise of cultural homogeneity within countries and the premise of cultural stability over time-both false in her view.Sivakumar and Nakata [49] noted that the samples in studies using a composite index of Hofstede's dimensions often pose methodological problems.For example, countries are sometimes too similar culturally to allow significant variations to emerge.
The GLOBE research project [50] is likely the boldest response to Hofstede-inspired research.It proposed a rival measurement tool using Schein's [1] cultural schemes.Although this approach is not completely incompatible with Hofstede's, it took aim at two basic premises of studies in the Hofstede mode: (1) the premise that measurement of individual values allows to draw conclusions on collective values; and (2) the idea that knowledge of a society's dominant values necessarily leads to the capacity to describe what is really going on in that society.Commenting on both research projects, Smith [51] noted that the GLOBE measurement tool made it possible to aggregate individuals' opinions on what they felt was best for society-a look from the subject toward the collective-whereas Hofstede's approach uncovered self-reported opinions on what the subject thought was good for him-a look from the subject toward himself.Each camp tried to impose its measurement tools and maintained that its number of cultural dimensions (five by Hofstede and seven by GLOBE) was more representative of reality.Many observers of this debate have remained sceptical and proposed to open completely new avenues of research.Earley [52] argued that analysing a complex concept such as culture with fragmented variables, no matter how many, was a dead end.His prescriptions were very clear: "(1) Stop doing this type of grand-values assessment; (2) Develop some mid-and/or grand-level theories that link culture to action; (3) Develop alternatives to values as a basis for exploring culture in relation to action."Maseland and van Hoorn [53] also noted that surveys and questionnaires, so frequently used by Hofstede and his followers and by GLOBE and its partners, were not methodological tools well adapted to understand the complexity of culture.In one of the most thorough reviews of the literature, Kirkman, Lowe and Gibson [54] concluded, after studying 180 articles and book chapters that used Hofstede's dimensions in empirical research, that this field remains highly fragmented.Many studies were repetitive, questions remained infinitely more numerous than answers, and it was not at all unthinkable, they added, that the time had come to go beyond Hofstede's foundational approach.Indeed, the functionalist paradigm takes a great risk in addressing culture by means of an almost exclusive focus on the concept of value and a search to establish various causal relations between culture-generally posed as an independent variable-and strategic choices.It also offers few means to understand whether the cultural variable is consciously considered by managers before they make their strategic decisions or is considered by them only when it emerges as they evaluate the results of their strategic decisions.All this no doubt explains the contention, so frequently expressed, that research emanating from this paradigm, despite its interest, remains curiously abstract and disembodied when one tries to shed light on the concrete behaviour of managers.

The Interpretive Paradigm
This other major paradigm in research linking culture and management finds its origins in ethnology, cultural anthropology, sociology, phenomenology, and semiotics.Many of these studies came from the non-anglophone world.In Le phénomène bureaucratique, French sociologist Michel Crozier [55] made one of the first systematic attempt to see how a national culture concretely influences management.He conducted an in-depth investigation of two public French organizations-the Parisian agency "Chèques Postaux" and "SEITA", a state monopoly for tobacco and matches.He asked whether the behaviours that he uncovered were common to all bureaucratic organizations, regardless of country where they exist, or were based on specifically French traits.Crozier was interested less in discerning different types of national culture than in highlighting the relevance of differences among nations.In L'acteur et le système, Crozier and Friedberg [56] then developed a phenomenological perspective based on comprehension of the subject's point of view.Positing that culture was a highly complex reality, they ultimately concurred with Geertz [57], for whom the study of culture should be a qualitative search for meaning, not an experimental science looking for universal laws.
Philippe D'Iribarne also marked out this interpretive paradigm.In La logique de l'honneur [58] and Culture et mondialisation [59], he attempted to understand the behaviour of groups of workers from various countries by referring to the historical, sociological, and philosophical origins of what he was observing.In his wake, a number of researchers [59] pursued his project of inventorying national management cultures and introduced various inflections.For example, whereas D'Iribarne was concerned with inventorying different national management logics, Dupuis [60] was interested in comparing these logics.He felt that shedding light on the dominant values of workers in a society might make it possible to compare and interpret differences and similarities between styles of national management.He identified blocs of influence issuing from Latin, Anglo-Saxon, and Nordic cultures at the core of the dominant management culture in Quebec [61].
Works based on distinct methodological approaches might also fall within the general interpretive paradigm.Morris et al. [62] studied how organizational actors situated at the juncture of different national cultures perceive a situation to be fair or unfair.Philips and Brown [63] used hermeneutics to study communications within and around organizations.Molinsky [64] studied the influence of emotions on the adaptation process of those who take part in intra-firm intercultural exchanges and advocated for input from psychology.Schein [65] also emphasized the usefulness of psychology in describing and understanding cultural conceptions implicit in the representations that humans have of an organization.Even linguistics made an appearance in Brannen [66], who showed how difficult it may be for a corporation such as Disney to preserve and transpose the image and values of cultural harmony when it enters a new national environment.
Undeniably, the complexity of phenomena studied by researchers using the interpretive paradigm often forces them to focus their research interest as tightly as possible, which isolates them from their colleagues.They also published less, often writing books rather than journal articles, and the concrete managerial relevancy of such research, though fascinating, might not be obvious.We share nonetheless the point of view of Birkinshaw, Brannen, and Tung [67] who credited qualitative approaches with providing access to a field that has received too little attention: the micro-processes at the core of the culture-context interface.

A Theoretical Framework to Access the Cultural Black Box of Strategy Formulation
As we have noted above, neither stream truly tells us whether the cultural variable is deliberately considered by managers during the formulation of strategy and its subsequent implementation, and whether they consider it when they examine the results of these strategies.It is as if these studies have sought to identify and measure a phenomenon of which they are not trying to understand the working mechanisms.This leads us to propose and test a way of opening this black box.
Our initial assumption was that disciplines such as philosophy and political science might contribute to our analysis of the role of culture in the formulation of business strategies.We saw promising avenues in the work of the Canadian philosopher and political scientist Charles Taylor (born 1931).Taylor's work is wide ranging, dealing with Hegel [68,69], the epistemological foundations of the social sciences [70][71][72], the emergence of Western modernity [73] and multiculturalism [74].We believed that Taylor's work could provide an innovative theoretical basis for the comprehension of internationalization strategies for firms whose leaders are conscious and sensitive of cultural difference.His work devoted to explaining the sources of modern Western identity contains a careful conceptualization of the foundations of the attitudes and practices of individuals when they are confronted with cultural differences.
Five concepts are at the core of Taylor's conceptualization: (1) positive liberty; (2) social ontology; (3) expressivism; (4) civic republicanism; and (5) common spaces.We propose to use these five concepts to assess and to interpret executives' discourses and practices in order to see how their vision of the culture variable influences their strategic choices.In other words, transformed into indicators that can be referenced, we posit that the following five concepts drawn from Taylor's work may become keys with which to open the black box of decision-making and see how cultural differences are conceived and managed by the people in charge of formulating corporate internationalization strategies.

Positive Liberty
The Taylorian conception of liberty is based on the work of Isaiah Berlin (1907Berlin ( -1997) ) and G. W. F. Hegel (1770Hegel ( -1831)).From Berlin [75], Taylor drew the notion of the difference between the "negative" conception of liberty of the subject (one is free who does not suffer external constraint) and the "positive" conception (one is free who assumes mastery of himself).Taylor favoured the second conception: "Doctrines of positive freedom are concerned with a view of freedom which involves essentially the exercising of control over one's life.On this view, one is free only to the extent that one has effectively determined oneself and the shape of one's life.The concept of freedom here is an exercise-concept.By contrast, negative theories can rely simply on an opportunity-concept, where being free is a matter of what we can do, of what it is open to us to do, whether or not we do anything to exercise these options" [72].
Hegel left Taylor with the idea that the quest for individual liberty is part of a broader historical movement of collective emancipation.Applied to organizational management of cultural difference, this idea helps to signpost whether the individual's quest for self-realization, in the context of a gap between his cultural identity and the dominant culture of his society, finds enough space to deploy itself or tends to be constrained.

Social Ontology
The collective component of individual identity is the second basic pillar.Taylor reminds us that "what has been argued in the different theories of the social nature of man is not just that men cannot physically survive alone, but much more that they only develop their characteristically human capacity in society" [72].Therefore, for Taylor, the subject's identity involves a constant coming and going between the collective and individual spheres.This is why Taylor accords particular importance to the social components of individual identity and distances himself from a strictly instrumental and individualist conception of rights and freedoms.Reformulated from a more practical angle, the concept of social ontology supposes the existence of a gap between the individual identity of a person and that person's social, especially cultural dimension.It is the acknowledgment of this division and its source that were sought in the discourses of executives that we examined.

Expressivism
In Taylor's work, the notion of expressivism refers to the individual's need for expression.Taylor posits that each person has an inner voice telling him that the truth is within him, particularly in his feelings: "my claim is that the idea of nature as an intrinsic source goes along with an expressive view of human life.Fulfilling my nature means espousing the inner élan, the voice or impulse" [73].If he feels legitimized by this self-confidence, the subject will want to make himself heard.This notion allows for an assessment of whether expression of difference is perceived as an outstretched hand encouraging the individual to enter in a dialogue with those around him, or if speaking out enables him to state his desire to live in this context of cultural difference.

Civic Republicanism
Like many before him, Taylor reminds us of the virtues of active citizen engagement in public life: "The struggle for recognition can find only one satisfactory solution, and that is a regime of reciprocal recognition among equals.Hegel follows Rousseau in finding this regime in a society with a common purpose, one in which there is a "'we' that is an 'I', and an 'I' that is a 'we'."( [72], quoting [76]).By becoming involved in public affairs, Taylor maintains, the individual contributes to the edification of the common good and ends up identifying with it [77], thus establishing harmony between the private and collective components of his identity.In a context of cultural difference, this notion may be used to evaluate whether an executive and his organization seek to fulfil their duty as a "good citizen" in the community.Obviously, this is also the concern driving much of the research on corporate social responsibility.

Common Spaces
Taylor regards social spheres in which the common good may be built as very important.He values sites of living together propitious to disinterested civic actions: "We can speak of 'common space' when people come together in one act of focus for whatever purpose, be it ritual, the enjoyment of a play, conversation, the celebration of a major event.Their focus is common, as against merely convergent, because it is part of what is commonly understood that they are attending to the common object, or purpose, together" [70].
Participation in democratic institutions-the political sphere-and volunteering-altruistic social engagement-are two typical examples.In our case, this notion can be used to assess the importance that a manager and his organization give to these spaces of intercultural dialogue and construction of the common good.
These five concepts will be at the centre of our investigation of internationalization strategies.In particular, we hope to be able to show: (1) that they are key elements in explaining how culture intervenes when expanding in foreign markets; and (2) that there is a value in using a grid inspired by the work of a philosopher to investigate the role of culture in internationalization decisions.Let us now explain the methodological foundations of our investigation.

Empirical Setting
We decided to focus on firms from the food processing industry since the food sector is intrinsically linked to national culture [78].A convenient sample was drawn to ensure the highest possible level of access and the cooperation of executives, which seemed essential in the context of research aiming for a dense interpretation of a phenomenon using a methodological strategy of case studies.Three criteria were used to establish a short list of multinational firms.We were looking for: (1) firms with processing facilities in three or more countries; (2) firms producing a broad range of products designed for the targeted national market; (3) firms showing sustained community involvement.We admit openly that a fourth and very practical aspect contributed to give shape to the list of forty companies contacted: personal networks of researchers.Four firms accepted to participate.It was agreed not to reveal their identities, their precise sales figures, or certain other details that would designate them too clearly.
Firm 1 was a Canadian multinational company with between 8000 and 10,000 employees and sales figures of CA$4-6 billion that distributes food products in 50 countries from its facilities in half a dozen countries on three continents.Listed on the stock exchange, this firm was founded by an entrepreneur who immigrated to Canada after the Second World War.Since its creation, the firm has woven very strong links with its host society by offering food products that have helped to popularize "ethnic" foods in Canadian everyday cuisine.
Firm 2 is a France-based multinational organization.It employs between 3000 and 5000 people in a dozen production sites in a half-dozen countries on two continents.Now listed on the stock exchange but at its origin a family business, it generates sales figures of between €500 million and €1 billion.It is a market leader in its home country and a strong presence in the community at both the regional and national levels.Beyond its national borders, it has adapted its production to meet local needs while offering a range of products that evoke its French origin.
Firm 3 is a privately held family multinational company headquartered also in France.It has between 3000 and 5000 full-time employees with sales between €500 million and €1 billion.Although a leader in France, its expansion toward international markets is more recent than for the other firms.However, it has undertaken production in three European countries and exports to about a dozen countries.
Firm 4 is a multinational company employing between 8000 and 10,000 people with its head office in France.Its sales figures are €1-3 billion thanks to sales in more than 50 countries and production facilities in a dozen countries on three continents.Its product lines are highly diversified and designed to match the specific demand in local markets.

Data Collection
Our data collection approach involved multiple informants in each firm (see Table 1).CEOs (at the first contact) and other participating managers (during their respective interviews) were asked to identify colleagues from any hierarchical level involved in decisions with regard to internationalization.This allowed us to select five respondents in one multinational, eight in a second, three in a third, and four in a fourth, for a total of 20 respondents.Three-quarters of identified participants were top executives: four were CEOs/Presidents and 11 were members of executive committees.Only five identified participants were working at a lower level of management and not members of the executive committee: four were at country/continental levels and one was an R & D senior manager.This sent us an early indication of senior managers' perception of the hierarchical level of decision making related to internationalization strategies.It was decided to respect this list of 20 senior managers identified by peers, with whom we conducted a total of 20.8 h of interviews.

Members of Executive Committees Lower Level of Management Total
Firm Interviews relied on a three-section interview guide.The first section was devoted to uncovering strategies and their determinants: general strategies, internationalization strategy, business environment, internationalization history, and factors in the choice of location.At this stage of the interview, the interviewer did not reveal his interest in the link between culture and strategy.Only in the second section, devoted to the attention paid to cultural difference, was the respondent able to glimpse the general outline of the research.At this point, the respondent was asked the following three questions: (1) "Can you explain the word culture?" (2) "Can you explain the expression national culture or, in other words, a culture specific to a country or nation?"(3) "Can you explain the concept of cultural difference?"The discussion was directed toward the influence of national culture on the firm, and then toward the formulation of internationalization strategies.The interviewer then revealed his interest in the notion of national culture and in understanding its influence-if there was one-on the firm.Finally, in the third section of the interview, business environment scanning practices were explored from three angles: (1) the information collected in the firm about the business environment; (2) executives' relations with this environment; and (3) the question of economic incentives encouraging attention to the international environment.The interviews were recorded and transcribed.
We also studied the firms' annual reports (a total of 15 for three firms over five years; the fourth firm is privately owned), 46 press releases from 2005 to 2010, and their Web sites.Finally, for triangulation purposes, we gathered 109 newspaper articles on these firms in 30 Francophone newspapers via the Eureka database.Consistent with Yin [79], varying documentary sources in order to corroborate and augment evidence is deemed important in a case study context.

Data Analysis
The first step in data analysis was to test the interview guide and procedures to validate the relevance of the five Taylorian criteria as analysis tools.A pre-test was conducted in a Canadian nongovernmental organization with the main mandate of attracting foreign investment to the country.The coding grid was then constructed to operationalize each of the five criteria.For each of the codes representing a criterion, three details were added: (1) definition of the criterion; (2) illustration and application of the criterion in a context of cultural difference; and (3) enunciation of the criterion indicators.Table 2 shows the indicators used in our discourse analysis to identify presence of Taylorian criteria.Using the coding grid, the interview notes were integrated into the NVivo 8 qualitative analysis software.The text was then carefully coded-that is, codes were assigned to each sentence or sentence fragment corresponding to an indicator in the criterion.One sentence or fragment might therefore receive more than one code, whereas another might receive none.The pre-test not only confirmed the relevance of the Taylorian coding grid, but data collected also revealed emerging themes and concepts we were not expecting to capture.Therefore, in addition to the coding grid for Taylorian cultural indicators, four complementary coding grids were developed to identify: (1) the firms' internationalization phases; (2) the level of executives' theoretical comprehension of the concept of culture; (3) the cultural factors influencing internationalization strategies; and (4) the other location factors influencing strategies.
In terms of unit of analysis, it was the firm's corporate internationalization strategy that interested us and not that of one or another of its divisions.Why?Because we hoped to reveal a coherent vision of the organizational strategy using an analysis of the discourse of senior executives and the organizational documents produced under their supervision.As highlighted in the Cambridge Handbook of Strategy as Practice [80], dedicated to the homonymous research movement, we believe that the discourse analysis methodology is a powerful tool to advance our understanding of strategy and strategizing.Because qualitative research is not devoted to uncovering causal relations, we chose not to identify dependent and independent variables.The coding and analysis were aimed, rather, at uncovering whether it is possible (and to what point) to identify and accurately describe the role played by culture in the formulation and implementation of strategies through the normative evaluation tools developed.
With regard to classic external validation procedures, our choice to proceed with a series of case studies opened the doors to this option.As internal reliability is also vital, we had to verify whether other researchers would arrive at approximately the same results by coding the same data.A second coding of the Taylorian criteria was therefore executed.This coding was performed by a PhD student with a management specialization at HEC Montréal.First, a list of 168 interview excerpts that had been coded using a Taylorian code was compiled and numbered.Second, a blind selection of 40 of these excerpts was made by four random draws (10 excerpts per firm studied).It should be noted that some of these excerpts had more than one code.Using the results of the senior coder as a reference base, the concordance rate was 88%.The calculation of the Cohen's kappa-taking into account the play of chance-also confirmed the reliability of the results obtained.The 40 excerpts that were double-coded were dissected into 47 mutually exclusive elements, referring to 38 interview elements that received identical codes and 9 others that received different codes.This exercise made it possible to identify a coefficient of 0.66, a level that Landis and Koch [81] view as substantial in their graduated evaluative scale.The changes made to the methodology following the pre-test were conclusive.It is therefore possible to claim internal reliability of the Taylorian core theory of this research.

Results
Our findings allow us to posit that the Taylorian indicators bring indeed to light the role played by culture in the formulation of internationalization strategies.Table 3 shows some of the observations collected for each indicator by firm.Firms 1, 2, and 4 proved to be very fertile ground for the Taylorian theoretical framework.Either four or five of the five Taylorian criteria were observed in these firms.Executives and their organizations seem to show notable sensitivity and savoir-faire with regard to management of cultural difference.In firm 3, only two Taylorian indicators were identified: social ontology and republicanism.Not surprisingly, this was the least internationalized of our four firms.
As illustrated in Table 4, the indicators of social ontology and positive liberty were the most frequently observed in all four firms, followed by the expressivism and civic republicanism indicators.The common spaces indicator was seen more rarely.Overall, these observations lead us to three key findings related specifically to the role of culture in general, the importance of national culture, and the quest for "cultural comfort" in internationalization decisions."with a bit more than 3000 employees, the owning family his more than aware of it social responsibility in the headquarter area." Minimal presence of this criteria in Firm 3 through limited participation into some health and safety regional committees.

Firm 3 No observation
"We said: we will settle, we will try to understand how they live, what is their culture, with a view to adapt ourselves."

No observation
An executive mentioned how Firm 3 contributed to the creation of an economic stimulus regional fund after the closure of a factory.

No observation
Firm 4 "Firm's workers are the most important resource of Firm 4." Success of strategies is based on employees capacity to seize, promote and to implement policies needed for their accomplishment." An executive explained that the new GM national origin has seriously been considered: "We have this opportunity to be located at the crossroads of three great cultures… When I chose the new GM (a North American), it was notably to bring rigor, rationality and common sense." Low level of presence of expressivism.However, Firm 4 annual reports mention the importance of "mutual respect", "to listen each other" and "to be available for other".
Strong demonstration of civic republicanism through: (1) sustainable development; (2) energy efficiency of factories; and (3) creation of a foundation dedicated to the promotion of healthy nutrition.
Limited presence of the common space criteria.However, an executive is very involve on the regional economic scene.

The Central but Informal Role of Culture
Observing the marked presence of Taylorian cultural indicators is an important step, but it does not make it possible to declare that the executives consciously implemented these specific practices of managing cultural difference.In firms 1, 2 and 3, sensitivity to cultural difference was certainly observed but we cannot establish an obvious link between such sensitivity and its deliberate use in the construction of an internationalization strategy.It is one thing to observe a particular sensitivity and another to know whether it is the result of deliberate planning or if it emerged spontaneously [82].

The Importance of the National Cultural Sphere for Multinational Firms
In three of the four cases (firms 2, 3 and 4), the national cultural sphere of the targeted market proved to be an important aspect of executives' interpretation of the business environment.Nevertheless, the organizational sphere, strongly influenced by the respective founding families of the firms we studied, dominated the intellectual horizon of executives in those firms.We found particularly interesting how the organizational sphere was reconciled with other sources of cultural influence.Three out of four firms placed such influences at the core of their interpretation of their business environment and looked for ways to reconcile them with their corporate identity.Of course, the fact that the questionnaire gradually revealed the researcher's interest in the national cultural sphere may have influenced its very strong presence in the respondents' discourse.The case of firm 1, whose respondents denied more than once the importance of the national cultural sphere, illustrates that we had nevertheless left the space necessary for the expression of such dissent during the interviews.Table 5 (below) illustrates our assessment of the influence of various cultural spheres on firms.

Firm 4
Dominant organizational sphere not strongly influenced by founding family: Firm 4 CEO explains that, even if the founding family is still present, "culture generally refers to culture of the firm".
Influence of various national culture is recognized and the objective of establishing linkages with the organizational culture is clearly present: " I note that Europe is still an amalgam of several countries with large differences (…) We have to adapt to these various markets" Food processing industry culture: -continental -regional -linguistic

The Quest for Cultural Comfort as a Factor When Choosing Location
The listing of location factors most frequently mentioned enabled us to confirm that when they seek to enter a new country, executives seek a comfort zone (a fit) on the cultural level by examining the national culture of the country targeted.This confirms the abundant empirical research already available [3].When we asked questions to force isolation of each location factor, the search for a comfort zone with the national culture was mentioned in 12% of the responses in which executives explained why they had chosen one country rather than another.That location factor was surpassed only by access to the market (16%) and closely followed by access to raw materials (11%).The search for a culinary fit came fourth (10%), on an equal rank with opportunity.When combined, the national cultural fit and the culinary cultural fit (22%) were the top-ranked location factor for three firms.We cannot state definitively that our firms were all seeking, first and foremost, such a combination before engaging in an internationalization project, as our methodology does not to enable us to come to that conclusion.It is however possible to state that these two cultural spheres were fundamental components of executives' attention in all four multinationals when they had to decide to enter one country rather than another.

Discussion
These findings triggered a number of further reflections for understanding the role of culture in internationalization.They underline what we believe could be key aspects of how managers might address cultural differences.

Management of Cultural Difference is Closely Linked to the Internationalization Stage
It also appeared that there was a likely link between a firm's internationalization stage and the intensity of its management of cultural difference.Firm 3, where our Taylorian indicators were weakest, was the least advanced in terms of internationalization of its activities.It performed processing activities in only two countries, whereas the other firms had processing activities in six or more countries.Their internationalization histories were also more advanced.We believe that this was not a coincidence and that the more managers compose with cultural differences, the better equipped their managerial tool box is, which is then reflected by a greater presence of Taylorian indicators.Nonetheless, we acknowledge that the link between a firm's internationalization stage and the intensity of management of cultural difference ought to be much more scrutinized if one hopes to go beyond a chicken-and-egg conundrum.

Culture Management is an Incremental Learning Process
Overall, the links between culture and management appeared to be informal, unconscious, and intuitive.In short, although the cultural variables were very obvious, it is not possible to state that the executives paid them conscious attention when they formulated and implemented internationalization strategies.Management practices for cultural difference appeared, rather, intuitive and were not crystallized in the form of an organizational strategy except as an emergent process as defined by Mintzberg and Waters [82].This suggests that the executives' aptitude to integrate various cultural variables into their strategic decisions is an incremental learning process.Again, this confirms existing findings, most notably those based on knowledge or resource-based views [83].Furthermore, in all four firms, decisions linked to internationalization were made by a small group of individuals, all working at the highest level of the corporate hierarchy, thus giving key importance to the issue of transfer of knowledge when departures happen.

When Planning International Expansion, Search Explicitly for Cultural "Fits"
In all four firms, internationalization strategies mobilized a basket of cultural "fits" of various types (national culture, culinary culture, professional culture, organizational culture, and so on).That might serve as a reservoir of potential means for distinguishing an internationalization strategy from those of competitors.Also, while our findings are exploratory and would need to be substantiated by additional research, we suggest that executives might gain from searching more explicitly for cultural and organizational anchors when they review location factors.We noted that some executives implicitly searched for these successful marriages, but without addressing and discussing them openly.It is certainly understandable that executive committees would discuss more tangible location factors-size of market, cost of raw materials, presence of a complementary industrial cluster, and others-as a priority.It would also be profitable to address the more informal factors more frankly and engage in discussion to evaluate them in a group context.
The fact that management thinking and practices related to cultural difference are intuitive, informal, unconscious and integrated through an incremental learning process raises a massive question: what role does intuition play in determining strategic choices having to deal with cultural differences?Or, to put it otherwise, is there a discernible pattern for transforming intuition into action?Let us be honest: our data does not allow to answer that fascinating question.It would require another completely different research design.One possible starting point could be the recognition-primed decision model proposed by psychologist Gary Klein [84,85], though it is true that Klein's approach was originally designed to understand quick decisions (firemen, nurses, etc.).Klein found that experts having to make decisions under high uncertainty and heavy constraints rely heavily on their experience to recognize situations similar to some already encountered in the past and then select feasible options.Intuition acts as a pattern-matching process.A chess player recalling past games and using them to determine feasible courses of actions, especially when playing numerous simultaneous games, would be a classic example.
In terms of methodology, the Critical Incident Technique or CIT, suggested by Acinki [86], and the Verbal Protocol Analysis (VPA), proposed by Baldacchino and his colleagues [87], could be two conceivable and fruitful tools available to anyone wanting to explore that avenue.CIT is based on interviews centered on the in-depth recollection of a key past moment experienced by the respondent.VPA is based on instructions to "think aloud" while acting and simultaneously making a detailed record of the person's verbal report, to be analyzed later.Both methods would need to be adapted to the particular context of business managers.It would also be crucial to take into account the classic warnings by Kahneman [88] and Tversky and Kahneman [89] on how intuition fuelled by overconfidence often carries biases which lead to illusions and mistakes.

Conclusions
By choosing to contribute to the interpretive paradigm using the method of normative evaluation, we were able to provide an in-depth description of a specific phenomenon.The flip side of the coin is that we were unable to observe and identify causal relations.We accept this limitation because it is not so much the capacity to generalize that interested us, but the experimental attempt to open the black box of decision-making.We noted previously also that management practices for cultural difference appear intuitive rather than conscious as existing research has already shown.This would tend to indicate that our methodology based on 1-h-per executive interviews might have been enriched by the use of a participant observation method.Furthermore, since the interviewer-interviewee relationship is a social one, involving, as noted by Brinkmann and Kvale [90], an implicit asymmetrical power relation between the person "owning" the information and the interviewer "needing" the words in order to extract usable data from them, its outcome can be potentially biased by various factors related to age, race, gender, social status, cultural background, difficulty of obtaining approval for the interview and other factors too.Did we do enough to minimize those potential biases?One can always do more, but since our method was highly systematic, it generated an external reliability which we feel helped minimize the possible impact of these biases.In retrospect however, the use of a reflexive journal, in which the interviewer reflects and writes, after each interview, on how he might have inadvertently influenced the process, could have been a useful additional tool.
It goes almost without saying that other theoretical approaches could have been mobilized and would also have yielded highly interesting results.The broad field of practice theories, for example, brilliantly and thoroughly reviewed by Nicolini [91], is basically grounded on the study of what people actually do and views behavior as largely composed of routinized and recurrent processes.It offers a wide array of approaches to the study of a topic such as ours.For instance, the work of French scholar Michel Foucault [92,93] is deeply suggestive to anyone interested in studying discourse.His focus was less on language itself than on the conditions and rules that give birth and shape discursive practices, viewed as largely contingent, which support the exercise of power.Anthony Giddens's theory of structuration [94], focusing again on praxis-what people do more than what they say-and viewing practices as the key link between actors and structures would have also shed light on how executives manage firm entry in a culturally different market.Another useful and interesting theoretical perspective could have been Bourdieu's ethnographic focus, specifically his notion of "habitus", defined as "(…) principles which generate and organize practices and representations that can be objectively adapted to their outcomes without presupposing a conscious aiming at ends (…)" [95].
The choice of industry, the size and provenance of firms, the type of shareholders, and many other variables also conditioned our results.The cultural dimension would also have looked radically different if we had studied its role in an entry strategy in a non-Western market or in a developing country.But this would require a larger team of researchers aiming to gain access to a greater number of more diverse firms.For such an undertaking an as stated previously, personal networks of researchers would appear to be essential to convince a higher number of firms to participate in such a project.
We can think of many extensions to our research.We could apply the same approach to a sample of firms in a different industry.Since corporate strategy and headquarters issues were the sole focus of our research, one might also want to study division-level decisions or how the culture issue is dealt with in the overseas subsidiaries.It would also be of great interest to involve a broader group of managers than the 20 senior managers identified by peers.In a paper dedicated to the narrative turn in organization studies, Fenton and Langley [96] suggest that strategy as practice could be seen as a "multi-actor, multi-level process of actions and interactions between actors making sense of their role and the roles of others in relation to the carrying out of strategy" [96].We agree that, even if not included in the more formal executive committees' decision process, first level and mid-level managers are likely to inform corporate strategies.
From a methodological point of view, the inclusion of lower level of management would have the benefit to increase the validity of results of future research and to offer a larger population and sample size which would allow using quantitative methodology.A larger scale cross-cultural research project would permit studying causal links and eventually allow generalizing observations.We would however have to pay careful attention to the design of the data collection and measurement tools and to the statistical modeling used to investigate the phenomenon.Steenkamp and Baumgartner [97] showed how important it is to assess the applicability of a single conceptual framework in cross-national research.In addition, Salzberger and Sinkovics [98] insisted on the importance of ensuring comparability of the data collected in different national populations, as different answers might be attributed to differences in behavior when responding and not toy real differences in the latent variables.It would also be crucial to differentiate between individual, hierarchical level of individual, organizational and cross-cultural differences to determine the origin of variations in behaviors and beliefs [99].
Another extension of this research project could be the introduction of a temporal dimension.We identified the prevalent internationalization stage of each firms at the time of data collection, but we did not study their genesis or evolution.Literature on organizational identity formation [100] and change process [101] reveals that social constructs, such as organizations or cultures, are not stable and that time and history are immensely important.It would be interesting to revisit our four firms in a few years and to interviews retired managers, to study how and what changes have occurred over time.As suggested by Malhotra, Peterson and Kleiser [102], historical methods and cultural anthropology could be of particular interest for researchers willing to conduct long-term research.
The link between the awareness/attention to cultural difference by top executives and the performance of our four firms could also be a fertile area for research.Finally, a critically significant dimension to explore would be to compare the explanatory power of our use of Taylor's five core concepts, which we believe was worth a try, to already existing theories in strategic management.Only then could we determine the place it truly deserves in the literature.

Firm 1 Firm 2 Firm 3
Dominant organizational sphere strongly influenced by founding family: "We are interested in things happening within our walls" (…) At the end of the day, firm's culture should prevail"."All decisions we are currently taking are guided by values inherited from the founding family".Minimized influence: " National culture or country culture, we don't spend too much time on that" No data Dominant organizational sphere: CEO's mission is to "transpose values, practices and management modes into each international division".---Recognition of importance of national culture of other countries: The CEO note "the difficulty and the importance of linking country cultures" No data Dominant organizational sphere strongly influenced by founding family: "The funding family is the majority shareholder of Firm 3. (…) This clearly means that (…) that we have a real DNA, a real identity, (…) Firm 3 culture exist and, for me, is fundamental".Important influence of national culture of original country Strong desire to adapt to different national culture: "Implantation certainly allow a better development, considering implantation means proximity and impregnation of culture".

Table 2 .
Indicators Used in Discourse Analysis to Identify Presence of Taylorian Criteria.
-Description of an identity-based (not necessarily cultural) difference -Explanation of the source of an identity-based difference -Identification of common features (without necessarily taking account of a specific sphere of influence)
-Creation of common spaces: with other interest groups, involvement in the cultural sector, involvement in the local sports scene, etc.

Table 3 .
Evidence of Taylorian criteria, by firm. HI

Table 5 .
Assessment and chosen examples of evidence of the influence of various cultural spheres on Firms.