Product-Country Images: Towards a Contextualized Approach

ABSTRACT - Country of origin has been seen as so important that it has even been suggested that Acountry equity@ should be a way of thinking about global brands (Shimp, Samiee, and Madden 1993). Yet COO is still a not well-understood concept, and mixed findings abound regarding its effects. Without a well-founded theoretical framework, the COO paradigm can neither address its limitations (see Bilkey and Nes 1982; Ozsomer and Cavusgil 1991; Papadopoulos and Heslop 1993) nor can it sustain generalisations about COO effects. Thus, we need to address conceptual issues concerning COOBits significance as a cue, how and under what conditions it is used, in order to understand the important implications for changing, maintaining, or modifying COO image.

Citation:

Soren Askegaard and Guliz Ger (1998) ,"Product-Country Images: Towards a Contextualized Approach", in E - European Advances in Consumer Research Volume 3, eds. Basil G. Englis and Anna Olofsson, Provo, UT : Association for Consumer Research, Pages: 50-58.

European Advances in Consumer Research Volume 3, 1998      Pages 50-58

PRODUCT-COUNTRY IMAGES: TOWARDS A CONTEXTUALIZED APPROACH

Soren Askegaard, Odense University, Denmark

Guliz Ger, Bilkent University, Turkey

ABSTRACT -

Country of origin has been seen as so important that it has even been suggested that "country equity" should be a way of thinking about global brands (Shimp, Samiee, and Madden 1993). Yet COO is still a not well-understood concept, and mixed findings abound regarding its effects. Without a well-founded theoretical framework, the COO paradigm can neither address its limitations (see Bilkey and Nes 1982; Ozsomer and Cavusgil 1991; Papadopoulos and Heslop 1993) nor can it sustain generalisations about COO effects. Thus, we need to address conceptual issues concerning COOBits significance as a cue, how and under what conditions it is used, in order to understand the important implications for changing, maintaining, or modifying COO image.

The application of stereotype and semiotic theory to research in country-of-origin or product-country-images is discussed. More specifically, it is argued that analyses of images attached to a product and its place(s) of origin must use a richer set of connotations and stereotypes than used in standard approaches, and that such images are context-dependent, i.e. must be understood in relation to the imagery of the local market and the usage or consumption pattern.

INTRODUCTION

Studies of the impact of a product’s country of origin (COO) have been a recurrent theme in conferences, journals and textbooks on international marketing and international business for the last couple of decades. However, this research has been largely a-theoretical and descriptive until recently (Papadopoulos, 1993). One recurrent theme has been the study of the disadvantageous image and position of products from less developed countries as opposed to products from highly industrialised countries. While this topic is certainly important for the study of the functioning of the world market, it has hardly ever taken us beyond the level of confirmation or relativisation of general stereotypes. Another much discussed problem has been the relative strength and weaknesses of Japanese and American products, an issue raised on the background of the trade relationship between these two economic giants.

Consequently, the orientation of the research has been rather practical. Not many efforts have been devoted to explore the fundamental questions behind such research topics such as: What is an image? How do images of product and place of origin combine to form a whole? How is an image dependent on its context?

In order to provide a satisfactory and elaborate answer to these questions, the necessary theoretical and methodological apparatus must be at our disposal. And, even though country-of-origin and related research has come of age, there are still important problems to be solved in the theoretical and methodological field.

Our central question thus becomes: How can we understand the relationship between the product image and the image of its place of origin? The issue here is to develop a conceptual framework and a methodology for the study of this relationship. That is, what are its dimensions and components? This question points to some possible extensions of the existing research in the field.

REVIEW OF EXISTING RESEARCH

The research in the effects of country-of-origin dates back to the mid-sixties. In the beginning, the research area focused solely on the single cue of the "made in" label, thus addressing a link between the country of origin and evaluation. The country of origin was thus used as a pure denotation without further specifications. In the first thorough literature review of the field, Bilkey and Nes (1982) argued that the research should be broadened to contain more than the immediate effect of the made-in label as the single cue studied.

To specify the made in label, later studies operationalised image as perceptions of products from a country: of overall quality, of attributes, of marketing and production properties of the country. However, empirical evidence soon indicated that such a general conceptualisation was not always valid in specific domains in that it did not depict the specific characteristics decisive for the evaluation of a particular product category. The country-of-origin research thus began first to focus on specific product images, then to combine both general and specific product images, stressing the necessity of such a narrow approach in order to predict buying behaviour (Papadopoulos 1993).

However, such an approach seems to restrict the concept of "country image". Consumers’ and buyers’ perceptions of countries’ products, as well as their feelings towards the people of these countries and the desired level and types of interaction with them were found to be aspects of country stereotyping (e.g., Bamossy, Berßcs, Heslop, & Papadopoulos, 1986). Such developments in the field suggest that to avoid the restrictedness of the narrow approach without missing the focus of it, and to capture the breadth of the eneral approach without loosing the connection to the specific product domain, a gestalt approach is needed, combining figure and background. What is meant by this is that the essence of the image is the whole, not the single parts since the interrelationships between the single parts contribute to the whole.

The first step in that direction is best exemplified by the work of Papadopoulos (1993). In a very comprehensive state-of-the-art article, the whole notion of "country-of-origin" is criticised as being misleading and too narrow, since it assumes a single place of origin for a product. It is assumed that the important origin for the end consumer is "manufactured in". While this is possible, it could also be also designed in, invented in, made by a producer originating from, etc. or a combination of these. Papadopoulos therefore suggests the broader term Product-Country Image (PCI) to account for the multidimensional character of the images of products/brands on the one hand, and of the multiple places potentially involved in a global production system with increasingly mobile products on the other.

The notion of product-country image may in itself be subject to criticism since it implies that the important place for the image construction is the country, understood as nation-state. But this may not always be the case. For instance, in terms of a regionalised conception of origin, the districts of Champagne and Cognac in France have been relatively successful in protecting their specific regional products from becoming generic product names used for products originating outside the region. From the tourist industry we know how places revert to highly distilled and simplified stereotypes (Belk 1996), which is evident, for instance, in Japanese tourists’ desire for omiyage gifts that capture meibutsu, or the essence of the place (Graburn 1987). [We thank one anonymous reviewer for providing us with this excellent example of what we are talking about.] Also, "made-in-Europe" is becoming a relevant label (Schweiger, HSubl and Friederes 1995). Thus, following this logic, one should ideally conceive of product-place images.

As already indicated, most of the research done in the past has focused on either the general country-of-origin image or the country-of-origin image of specific product categories. Possibly most influential among the few attempts to discuss this interface is the work of Han (1989). Here, a distinction is proposed between country image as a halo effect, where the general image of a country’s products is transferred to the specific product category, and country image as a summary construct, where the specific product images are the building blocks of the general image.

Papadopulos (1993) is also mentioning another shortcoming in traditional PCI research. Many researchers have focused too much on classical products and services and not considered other "product" types like whole countries, companies, etc. (see Ger 1991, Graby 1993, and Kotler, Haider and Rein 1993 for exceptions). If image is considered as that meaning system which we attach to a phenomenon, and we assume that meaning systems are used to interrelate phenomena, it follows that the image of one phenomenon is not independent from the image of other related phenomena. Thus, from a systems theory perspective we would expect that the PCI of a specific product category is directly linked to the general image of this country, not only indirectly through the general image of other products, as it seems to be indicated by the summary construct. In general, it can be said that the halo/summary construct approach remains on a very general and abstract level, avoiding the discussion of how different contexts or product categories may influence the imagery created.

Furthermore, the image of one product category from one country is not independent of the images of other related phenomena, such as competitors. Imagery depends on the relative position of the object among its competitors (Poiesz 1989). Thus, the notion of PCI, as any other image, needs to be built upon the broad context of competitive position, as well as the general image of that country. This need for re-conceptualistion is suggested by the notion of "image" itself, or meaning including linkages to country and competitors, and the marketing strategy of positioning.

That such a general relationship has an impact on product evaluations is consistent with the findings of several researchers (see Niss, 1994 p. 20-21 for a review). For instance, the scale developed by Pisharodi and Parameswaran (1992) includes general country items, general product items and specific product items. However, the scale cannot account for a very profound understanding of the imagery attached to a specific country, since it only captures a subset of possible descriptive dimensions of a country, all relating to socio-economic development. Furthermore, their a priori assumption that greater similarity between countries is positive is challenged by the globalising of economies and societies. It has been argued that much of the consumption occurring today is specifically consumption of culture, of the strange and the Other in the compressed form of specific goods, settings or events (Frat 1995), thus undermining the standard assumption in much COO literature that the more familiar, the better.

The context of existing production and usage traditions for a specific product category are hardly ever taken into consideration when studying PCI’s. In this respect, Morello (1993) presents an interesting addition to the notion of country image, in that he contextualises it in relation to the competitive field of the market in question. As one of the few scholars in the field, he thus relativises product-country image with respect to the market context. Although his discussion is somewhat restricted to alternative communication strategies adapted to the four possible situations in a two-x-two matrix of strong/weak competitive positions and favourable/non-favourable images respectively, the idea is a welcome opening of the concept of country-of-origin image to its environment.

A final problem that we would like to evoke here is the standard S-O-R communication approach normally used. In most COO literature, the stimuli evoking the country-of-origin are not problematised at all. For example, in the Pisharodi and Parameswaran (1991) scale the meaning of a country is supposedly contained in the industriousness, education and standard of living of a country plus its political, economic, and cultural similarities to the respondents own country. The meaning of these variables for the respondent is hardly ever discussed, i.e. the recipient’s cognitive processes are so to speak taken for granted in advance. It has been demonstrated elsewhere how the consumer’s reception of messages is in part independent of the coding of these (Mick and Buhl 1993). For an imagery as complex as the relationship between a country and its products (in the term’s broadest sense), we consider that such an assumption threatens the validity of the research.

To sum up, we have been witnessing important developments in the last couple of years in the country-of-origin research, of which the most important are the following:

1) The introduction of the product-country-image concept, which takes into account the complexity of the production and branding processes in today’s international markets.

2) The related understanding that the place of interest in this connection is not always one country, but may be several countries, a region, a continent or subcontinent, etc.

3) The contextualisation of the product-country image in a market setting.

Here, we would like to add a fourth which we have not been able to spot in the literature, but which we nevertheless hold as central for the theory development in thefield:

4) The inclusion of the "receiving subject" in the formation of the imagery in question, not only as an evaluator of otherwise objectively given information but as a creator of meaning in him/herself. The reader is, as maintained by reader response theory, "engaged in a dialogue with the text and a participant in meaning creation" (Stern 1989).

One concept that might be useful in capturing these developments and thus expanding our understanding of such images is the concept of stereotype, since this notion might introduce the context as well as the reading subject to the study product-country imagery.

STEREOTYPE THEORY: PCI AS MENTAL REPRESENTATIONS

Papadopoulos (1993) finds it unfortunate that stereotype research as known from social psychological and political research has not yet found its way into the field of product-country images. Consistent with the social construction of meaning approach, object schemata have been argued to be cultural schemataCclassifications/organisations of knowledge shared by members of particular cultures (Casson 1983). We start with the premise that image of productsCmental representations of categories of products, are similar to stereotypesCmental representations of groups of people.

Recent studies about COO or country image have, however, begun to consider stereotypes and related concepts such as schemas and cognitive structures (Ger 1991; Maheswaran 1994; Shimp, Samiee, and Madden 1993). Maheswaran (1994) suggests that COO is used as stereotypical information in making evaluations, and that COO represents a knowledge structure which links a stimulus to highly probable features. Ger (1991) argues for expanding the concept of COO to include the image of the country, and defines country image as a schema, or a network of interrelated elements that define the country, a knowledge structure that synthesises what we know of a country, together with its evaluative significance or schema-triggered affect. She argues that investigating the content of the country-schema, including associations with other countries, will add to an understanding of the role of country image on product evaluations. Such a broadening is consistent with the connotative meaning (including potency and activity as well as evaluations) and the symbolic meaning of any concept. It is also consistent with positioning being relative to competitors, and taking into account imageCsymbolic and intangible aspects of products. To build upon these suggestions, and move towards a fresh conceptualisation of PCI, we examine the nature of stereotypes.

The term "stereotype" derives from Greek "stereos", meaning solid, and "typos", meaning mark of a blow, impression, model. The term was first used by Lippmann in 1922 to refer to "pictures in our heads" that we use to apprehend the world (Seiter 1986). Stereotypes, of peoples’ concern, can "be conceived of as culturally shared categories that transcend the individual" (Stroebe and Insko 1989, p.29) and thus part of a people’s social heritage, widely shared within a society, and acquired through socialisation. They are culturally defined: Folklore (proverbial sayings, tales, songs, jokes) may play an important role in the creation of company and product imagery (Sherry 1984) and media are of course very influential in forming persistent stereotypes (Bynum 1990-1991).

Stereotypes are representations of groups, often used to describe, interpret, evaluate, and predict actions of individuals (Bar-Tal, et. al. 1989; Oakes, Haslam and Turner 1994). Stereotype effects include distortion of perceptionCbias in selectivity, accentuation of both differences between, and similarity within categories (in- and out-grops), perceived homogeneity of out-groups, confirmatory biasCtendency of stereotypic expectations to confirm themselves, and evaluative biasCevaluative and behavioural discrimination favouring the in-group. Below, we briefly review current approaches to stereotypes, and discuss how regarding PCIs as stereotypes contributes to the understanding of PCI effects.

The cognitive approach to stereotypes

The cognitive approach views stereotyping to be a simplification strategy employed by the perceiver to facilitate his or her interactions with a complex environment. Stereotypes are regarded to be shortcuts, and cognitive memory biases. Distortion of perception, expectancy-confirmation bias, and evaluative bias are viewed to be due to selective encoding and retrieval (of expectancy-consistent information), category label (stereotype) being used as an efficient cue for category instances, rehearsal of related items, and elaborateness of processing for incongruent information. Stereotyping involves the influence of category-based expectations on the processing of relevant information (Stangor 1989).

New information is seen with respect to such pre-existing mental representations. Categorisation allows one to relate new experiences to the old: once we assign an entity into a category on the basis of its perceptible attributes, we can infer some of its non-perceptible attributes (Medin 1989). Concepts are organised, creating coherent structures. This organisation involves similarity relations, in a broad sense: including attributes, relations, and higher-order relations, interdependent properties, properties at multiple levels of abstraction, and a depth and structure created by properties and relations (Medin 1989). Thus, all perception reflects categorisation (Bruner 1957), and categorisation is intrinsically comparative and relational, representing "the variable relations of similarity and difference in our environment" (Oakes, Haslam and Turner 1994, p.115).

The above views imply that similarity versus distinctiveness comparisons need to be considered in a conceptualisation and measurement of PCI. Such a conceptualisation needs to take into account similarity relations, in the depth and structure created by interdependent properties at multiple levels of abstraction, ie, within different kinds of interconnected and relevant mental structures. The notion of interconnectedness has been extended to include sensory and affective aspects and processes, and relevance refers to context, which activates specific goals, motives, feelings, and thoughts.

Conceptualisations of meaning representation have become broader, including sensory imagery and affect, along with cognition, with implications for stereotypes. A mental representation can involve interconnected verbal and pictorial aspects, operating in conjunction with a more general semantic memory (Marschark et al. 1987). And, it is now accepted that there is a continuous mutual interaction of cognitive and affective processes in memory (e.g., Mackie and Hamilton 1993; Stephan and Stephan 1993).

Categorisation in the memory implies that persons interpret and evaluate each object by comparing it to information already in the memory, along with its affective and sensory links/tones. Thus, stereotypes (hence PCIs) should involve cognitive and affective networks, with verbal and sensory/pictorial aspects, associated with the category. From now on we will use the term mental structures or representations to refer to interconnected cognitive, affective, and sensory structures.

Contextual nature of stereotyping: informational and motivational effects

Recently the role of context on stereotyping has been emphasised. Stereotypes seem to influence perceptions and evaluations in multiple, complex ways, depending on motivational as well as informational mechanisms (e.g., Bar-Tal et al. 1989). Both abstract and specific types of information seem to be used a needed: whether judgements involve prototypicality or exemplars, and elaborate versus simple, or top-down or bottom-up processes depend on contextual factorsCenvironmental cues, processing conditions, processing capacity, perceiver goals, involvement, motivation, and familiarity (e.g., Oakes, Haslam and Turner 1994). Findings also indicate that stereotypes do not always automatically lead to expectancy-confirmation (Stangor 1989). Prior expectations thus interact with other personal beliefs, and do not necessarily have to automatically influence judgements.

Hence, stereotypes are more than shortcuts and cognitive memory biases. Cognitive, motivational, sensory and affective processes combine in a given context to produce specific stereotyping effects (see also Hamilton and Trolier 1986). Categorisation enriches and expands the stimulus, allows a "more elaborated, connotative meaning" (Bruner 1957, p.148): "... categorisation works to maximise information ... by selectively drawing out aspects of structure, of similarity and difference in stimulus information, which are relevant to the perceiver’s current requirements within the stimulus context as a whole" (Oakes, Haslam and Turner 1994, p.113). The current COO findings may appear to be inconsistent due to the different contexts not being explicitly taken into consideration. In conceptualising PCIs multiple contexts must be considered.

Individual-group (self-object category) relations in stereotypes

Stereotypes embody symbolic values, those beliefs that are related to an out-group’s position relative to the in-group on deeply held ways of viewing and valuing the world (e.g., Mackie and Hamilton 1993; Oakes, Haslam and Turner 1994). This suggestion is consistent with stereotypes being a culturally defined and shared categorisation. Categorisation is necessarily an ordering, and ordering is not value-free. Stereotyping involves a differentiation between group and self, and others are categorised in relation to the categories associated with the self. Perception is relative to one’s own social and psychological place, in a comparative context (frame of reference). Finding that accentuation depends on the degree to which stimuli share the same category membership as the perceiver, Oakes, Haslam and Turner (1994) argue that stereotypes are sensitive to comparative context and reflect social self-categorical properties of stimuli. Stereotypes are thus related to ideological motivations and relationships to social and economic power, at a macro level (Seiter 1986). As Berger and Luckmann (1966) stress, any society’s ordering of reality is a historical product and is necessarily implicated in the power relations in that society.

These views are consistent with the earlier views that projection and defence are involved in stereotyping, and with similar more recent findings in international relations. For example, Volkan (1988) argues that images or representations of nations are formed partially by defence mechanisms such as externalisation and projection. How does individual-group relations of social stereotyping translate to PCIs? Social identity translates to identity through goodsCpeople express and define their identity through consumption (see e.g., Douglas and Isherwood 1979; McCracken 1988). Thus, regarding PCIs, we need to find out what the relevant "groups" for a particular product are.

TOWARDS A CONCEPTUAL MODEL

A mental representation of an object entails interactive cognition, affect, motivation, sensory, and verbal aspects of the meaning of the object, linked to multiple categories that inherently involve relational and comparative aspects. As categorisation is intrinsically comparative and relational, as these relations involve many levels since mental structures are interconnected at multiple levels o abstraction, as these structures include sensory and affective aspects, as a context specifies relevant structures, and as the person actively elaborates given his/her context-specific goals, motives, and values, an imageBa representation of a product involves all that depth, and all the comparative contextual relations.

However, as we have seen, current conceptualisations and measurements of COO or PCIs are inconsistent with the above views of mental representations, and are too restricted to capture the contextual and comparative nature of meaning. Although image, stereotypes, mental representations involve cognitive, affective, sensory, and motivational aspects, that go well beyond features or attributes, many COO studies use attributes in measurement, just like the prevalent use of traits in nation stereotypes. A comparison of quantitative techniques of measurement of nation stereotypes indicated that all measures elicited the same type of cognitive processing (Stephan, et. al., 1993). To date, COO studies have all had restricted cognitive perspectives and have not explicitly taken context (except stimulus factors such as type of product, level of technology, and cognitive perceiver factors such as familiarity) into account. The above review suggests that processing and effects of PCIs will be context-specific, depending on motivation, affect, and the symbolic values regarding the product category/country-consumer relationships. This calls for a non-positivistic, interpretative approach to such imagery.

We have argued that stereotyping is a fundamental categorisation tool and that stereotypes in relation to PCI must define the relevant categories used to evoke the meaning of the PCI. We suggest that product is a "member" of, or categorised in relation to, a product category, a set of competitors, a place, and a usage context where the usage context also captures the above-mentioned relationship of the particular product to "self". These dimensions basically account for the following fundamental questions pertaining to any object: what is it? what is its relation to other objects? where did it come from? and where will it go/what will it do? On the basis of these reflections, we thus propose place, product (or phenomenon), market context, and usage context as the four dimensions in a conceptual model of contextualised product-place images (CPPI), a term we would like to suggest as an at least temporary alternative for the PCI concept. Future research may hopefully come up with a less ugly term which better captures the comprehensive nature of the subject-matter.

Place refers to the meanings pertaining to all sorts of spatial relationships. This may be place of brand, place of production, place of design, place of innovation, place of origin of the raw materials etc. Needless to say, the discussion of the relevance of "country-of-origin" belongs in this sphere. But it also refers to meanings pertaining to the local and the global, the domestic and the foreign, and similar dichotomies. Indeed, the whole notion of origin is questioned by the multiple meanings attached to the potential multiple origins of a product.

The notion of product is here translated to phenomenon, in that we wish to distance ourselves from a narrow, material understanding of the things in question here. Phenomenon refers to "product" in that very general sense it has come to be understood within the marketing literature, i.e. any "marketable", tangible or non-tangible phenomenon. One of the basic questions here is, of course, the fundamental categorisation of the phenomenon in a foreign context. Is this product perceived as being the same in the domestic and foreign market?

The market context refers to the relative meaning of the phenomenon compared to its surroundings, competitors, etc. Relevant questions in this connection are such questions as: What kind of similarities and differences exist compared with phenomena (products) on the local market scene? And what kinds of distinctive features are used to demarcate and position goods in the market in question? This dimension points to te importance of positioning in the creation of a PCI.

Finally, the usage context refers to meanings pertaining to the consumption situation or to general and specific rituals concerning the consumption or usage of a given product. This, then, refers to the consumption context in general and the impact of origin spaces on the applicability of the phenomenon, and its compatibility with local usage scripts. (Figure 1).

Although stereotype theory may help us pointing out the contextualisation of images, it also has its limitations. Images as mental constructs are, so to speak, already there. This means that they are not really considered as elements of communication, ie, dynamic processes taking place in a given social and temporal environment. If PCIs involve an interrelated and embedded set of stereotypes or mental structures capturing all associations with a product from one place, and the goals, motives, and values of the consumer, and those of his/her socio-cultural group, we believe that the concept of image has been taken beyond the definition used by Poiesz (1989) and Ger (1991) as "holistic impressions" and given a dynamic element. Image, in this sense, becomes a text, a narrative, where the temporal dimension plays a crucial role. Therefore, we would like to go beyond the cognitive approach of the stereotypes and add, admittedly, a very few ideas from semiotics in order to better understand the narrative of the PCI. Here, we would like to underline that the use of structural semiotic models in the following does not mean that we adhere to structural semiotic theory in the classical sense, where meaning resides in the text itself. Rather, we will stick to our point from above that meaning formation takes place in the reader’s mind. However, this does not mean that the text does not provide some constraints to interpretation to the extent that draws on a common body of cultural conventions (Stern 1989).

FIGURE 1

DIMENSIONS IN A CONCEPTUAL MODEL OF THE MEANING STRUCTURE OF PRODUCT-COUNTRY IMAGES

THE IMAGE AS NARRATIVE: SEMIOTICS AND PRODUCT-COUNTRY IMAGES

"To each memorable image you attach a thought, a label, a category, a piece of the cosmic furniture, syllogisms, an enormous sorites, chains of apothegms, strings of hypallages, rosters of zeugmas, dances of hysteron proteron, apophantic logoi, hierarchic stoichea, processions of equinoxes and parallaxes, herbaria, genealogies of gymnosophistsBand so on, to infinity". With this quotation from Eco’s "Foucault’s Pendulum", Papadopoulos (1993, p.3) launches his article on "what product and country images are and are not".

Unfortunately, the use of Eco and other semioticians is not carried much further than that. In general, given that product-country-image research must fundamentally study the concept of meaning and meaning formation, there is surprisingly little reference to the field of semiotics as a source of theoretical and methodological input. The only study we have come across that at least partially uses a semiotic approach to the study of PCI is by Niss (1994). This study is addressing producers’ use of country-of-origin communication, not consumers’ or buyers’ perceptions of product-country images.

So while Papadopoulos (1993) does refer to Eco for a definition of the concept of image and a description of its functions, his recommendations for cross-fertilisation of ideas and findings among various disciplines surprisingly does not include semiotics. Yet, semiotics addresses the heart of the subject matter in image research: the social construction of meaning.

Semiotics, according to one of its founding fathers Swiss linguist Ferdinand de Saussure, is the science of the social life signs. It is the discipline which allows for a systematic analysis of symbolic and other sign systems, best introduced to consumer research by Mick (1986). A mental image (and, thus, PCI’s) has been defined as "a semiotic instrument necessary in order to evok and to think what has been perceived (Piaget and Inhelder 1971, p.381). Human beings use images for classifications and hierarchisations, for correlating phenomena in syntagms and deciding their substitutability in paradigms. We use images for symbolisations as in metaphorical constructions and to construct and express beliefs and explanations, thereby sometimes forming syllogisms and causal relations (Papadopoulos 1993 p.5). However, all these processes are processes of mind and subject to change, due to the fundamental ambiguity of the image. The image may draw its content from the world of realities, but its correlation with this world is always uncertain and subject to error (Berger & Luckmann 1966). Images, in short, are not our access to reality; they are, in this sense, our reality.

The contextualised product-place image is a narrative; it tells us a story. It tells us a story about the place of origin of the specific product (made in...) but perhaps also of the place of origin of the generic product (green-white-red colours on the pasta made in Belgium), the home of the brand etc. The product itself tells its own story according to its classification in the specific market context: Where is it located in the supermarket? Next to which other products? What do the advertising campaigns state? How does this differ from competitors’ and substitutes’ campaigns? And finally, it also provides meanings for the consumer pertaining to the effects of the usage. Will this food product be suitable for an exotic evening with friends? Or for a traditional meal? Will I be (regarded as) an efficient and skilled DIY person with a South Korean tool kit? And so on. Hence, the product plays a central role as a "potential" helper of the hero in one little "slice of life story-part" of the consumer’s own life, to paraphrase some standard elements from narratology. (Figure 2).

In the following, we would like to briefly describe, how three different semiotic approaches may contribute to throw light on the notion of product country image: Peirce’s triadic model of the sign, Barthes’ use of the concepts of denotation and connotation, and Greimas’ semiotic square.

FIGURE 2

A NARRATIVITY MODEL OF PRODUCT COUNTRY IMAGES

Peirce’s triadic model of the sign

A sign, according to Peirce, consists of three elements: The "Representamen" (sign vehicle), the "Object" (the sign vehicle’s referent), and the Interpretant, this being the mental image produced by the sign vehicle and its referent: "something created in the mind of the interpreter" (Peirce, cit.in N÷th 1995 p.43). Such an image, however, is not the end stage of the process since it is in itself a sign. This lead Peirce to consider the process of "unlimited semiosis", referring to the fact that any interpretation is in itself a sign which would be subject to interpretation and so on infinitely.

The process of unlimited semiosis may help to throw new light on the idea of associative networks, so prevalent in cognitive psychology, consumer research and also in certain parts of the country-of-origin research. The associations can be read as interpretants originating from the process of semiosis. This is not a linear process. The potential spreading of associations can go back, diverge, converge, etc. but due to the fundamental ambiguity of the sign, this associative process is not predetermined but always has several potential outcomes. Therefore, we do not normally end up in, e.g., a closed circuit in our thought processes. This has two consequences of central importance to us here:

Firstly, the complexity of the semiotic construct of a "country" and the open character of the Peircian sign relation makes it very difficult to establish the meaning of a "Made in" concept in advance, and it is consequently difficult to operationalise country-of-origin concepts with relatively short and closed types of questionnaires. Likewise, the cross cultural validity of conceptualisations pertaining to PCIs is called into question. Secondly, to the extent that certain lines f associations become predominant within a cultural setting, we may form stereotypes. Stereotypes can be regarded as normal reactions whereby "exigencies of practical life inevitably cut short the potentially endless development" (Gallie, cit. in N÷th, 1995 p.43) of the semiosis. So, on the one hand we have an infinite openness which makes the imagery concerning a phenomenon very difficult to entrench and, on the other hand, a cultural and individual imperative of cutting off the semiosis to form stable and relatively fixed images for categorising the physical and noospherical (mental) environment.

Barthes: Myth, Denotation and Connotation

One way to address the total image of a country is at what Barthes (1970 [1957]) called the mythical level, where what is at stake is "the second semiological system", the "meta-language" of the "total associations of a concept and an image" (op.cit. p.199-200). Although we are sceptical towards the Barthesian reduction of communicative material to "language", we do subscribe to the idea of image as a meta-level of signification systems (a configuration of interpretants, in Peirceian terms). Thus, a country and a nation is not only a mythical community in itself (Morin 1984) but may also engender mythological narratives in other countries as they become part of the signification process concerning "foreign-domestic" and "country X-country Y" (cf. the greimasian square below).

Myth, to Barthes, is a second order semiotic system which builds on the principles of connotation. In his early writings Barthes distinguished sharply between denotation as the primary sign, the immediately visible/audible signifier/signified, which in turn evokes a set of connotations, of second order meanings. However, he later modified this viewpoint, stressing that there is no objective reality in which to locate the denotative relation. On the contrary, he placed the denotation as the supreme mystification, an interpretation that seeks to hide its reality as such. This critique of denotation as an immediate reading was later developed much further by Baudrillard who, with reference to Barthes, stated that the "denotation is nothing but the most beautiful and the most subtle of the connotations" (1972 p.192).

Let us take the example of the statement "made in X". What does it denote? Even the denotation of "made in" is far from clear. As Papadopoulos informs us, "made-in can mean manufactured-in but also assembled-, designed-, or invented-in, and, often, wanting-to-look-like-it-was-made-in" (1993 p.4). Turning to the denotation of "X", it seems to be easier at a first glance: It denotes a place in space (and time!). At second thought, however, it is first of all far from obvious that it denotes anything at all, in that the country or its name may not be known to the interpreter. And even if there is a denotation attached to "X", it has no meaning apart from the sum of connotations attached to it. The "United States" as a geographical entity, would of course not be the "United States" prior to or independently of the human activities connected to that geographic area. Referring again to Morin, a country or a nation is in itself a mythological phenomenon; a second-order set of connotations.

The myth of the nation may be taken as a stereotype writ large. Stereotypes of nations and countries tend to make use of at lest three of Barthes mythological rhetorical figures (1970 [1957] p.238ff.): Denial of history (the stereotype is eternally valid), tautology ("it is what it is"), and assertion (the proverbial character of stereotypical logic). To give an example, Barthes (1970 [1957] pp.74-77) himself among other things makes a brief description of the French mythology of wine. This myth, as is the mythology of cheese and other products, is constitutive of the French nation, but it is also an inherent part of the stereotypical image of France, the French and French products abroad (Graby 1993). Obviously, this image is not static but wll change according to social and historical circumstance, as for instance when the French Government decides to disregard a large part of the world opinion and resume nuclear testing in the South Pacific.

FIGURE 3

GREIMASSIAN SQUARE

Consequently, neither the phenomenon ("product") nor the place ("country") can be taken at face value when we talk about PCI’s . They are signs whose meaning must be determined through the demasking of their denotations and the analysis of the connotations they evoke, and, hence, the myths they are inscribed in.

Greimas’ Semiotic Square

The preceding arguments raises the question of contextualisation. One way of addressing the contextual embeddedness of the meaning of PCI’s is through the semiotic square developed by Greimas. This model, simplifying Greimas’ very elaborate theory of the signification process, takes its point of departure in two oppositions: The contrariety between an assertion and its negation, and the contradiction between assertion and negation and non-assertion and non-negation respectively. The signification is created through these oppositions (N÷th 1990, p. 317-19). (Figure 3).

It follows logically that the meaning of such concepts as "foreign" or any more specific name of a country of origin gets its meaning in this oppositional context of what it is not. Such meanings pertain to the market context, to competitors and alternatives (paradigmatic relations) as well as to the usage context in which the product is inserted (syntagmatic relations). Thus, any meaning of a PCI must be contextualised in the market and the consumer culture holding that particular image.

Although one of the fundamental contrarieties in the analysis of PCI’s will often be the "foreign-domestic" contrariety, the situation may of course vary according to the market situation. In a non-car producing country, the fundamental contextualisation of the image may oppose the two leading exporters of cars to that country, e.g., Germany and Japan. Again, the important thing here is to assess exactly which contextual meanings, differentiations, and comparisons come into play in a given market situation. There is, as we have seen, an unlimited number of steps in the semiosis process which may add meanings from the comparisons and associations.

Semiotic analysis of images related to products and their places of origin call to attention the profound openness and ambiguity of the imagery involved, the importance of the mythologies attached to different parts of the world and the most fundamental questions of identity creation in relation to that, which does not belong to the identityBthat there cannot be a we without there being an other. Thus, semiotics points among other things to the importance of the imaging subject, to the dynamics inherent in the product-place imagery and to the multidimensionality and the context-dependency of such imagery.

CONCLUSION

Ideas of the open character of the sign, of the mythological foundation of our ideas of countries and nations, and of the signification process have, until now, not been used to analyse "product" and "country" meanings in a PCI context. We believe that a semiotic approach to PCIs would inscribe PCI research in a long and established tradition of meaning construction as well as add an important social and contextual validity to it. The more social approach of semiotics finds its individual counterpart in stereotype theories, and there may be a lot to gain from combining the two sets of theories. For instance, Mick (1988) used a combined semiotic-schema theory, akin to the approach suggested here, to analyse advertising meanings, using the Peircian triad to construct his conceptual model.

Drawing on theories of cultural and individual levels of meaning onstruction, based on semiotics and stereotyping we have tried to discuss some of the issues, explain the mixed findings, and provide a rich and fresh perspective to assess effects of place related images concerning and surrounding a product. We formulate such a framework to address if and how product-country images are used in product evaluations and choices. The approach proposed here combines stereotypes and semiotics to account for the roles of the social environment and the individual social psychology in the construction of images of phenomena, in our case the formation of product-country images. It thus reflects the interface between the individual and society, as it is conceived in social constructivism (Berger & Luckmann, 1966).

We have demonstrated how the two research traditions can be used both to criticise current research in the COO and the PCI tradition, and suggest a more comprehensive framework for the understanding of the work of the significations of linkages between products and places in the increasingly global consumer culture. We have suggested that four dimensions are central to the understanding of what we have termed the contextualised product-place image. Ongoing empirical research will hopefully demonstrate the strength and richness of such an approach, and help us to throw new light upon consumer imagery. And, most importantly, a first small step has been taken in the direction of inscribing imagery related to products and their places of origin in communication theory.

REFERENCES

Bamossy, G., J. Berßcs, L. Heslop, & N. Papadopoulos (1986), "East Meets West: A Country of Origin Analysis of Western Products by Hungarian and Dutch Consumers", Proceedings of the Second International Marketing and Development Conference, Budapest, 149-152.

Bar-Tal, D., C.F. Graumann, A.W. Kruglanski & W. Stroebe (eds) (1989), Stereotyping and Prejudice: Changing Conceptions, New York: Springer-Verlag.

Barthes, R. (1970 [1957]), Mythologies, Paris: Seuil, coll. Points.

Baudrillard, J. (1972), Pour une critique de l’Tconomie politique du signe, Paris: Gallimard

Belk, R.W. (1996), "Hyperreality and Globalization: Culture in the Age of Ronald McDonald", Journal of International Consumer Marketing, 8 (3 & 4), 23-38.

Berger, P. & T. Luckmann (1966), The Social Construction of Reality. New York; Doubleday.

Bilkey, W.J. & E. Nes (1982), "Country of Origin Effects on Product Evaluations," Journal of International Business Studies, 8 (Spring/Summer), 89-99.

Bruner J.S. (1957), "On Perceptual Readiness," Psychological Review, 64, 123-152.

Bynum, J.C. (Winter 1990-1991) "Folklore and International Stereotypes," Et Cetera: A Review of General Semantics, 47 (4) 399-402.

Casson, R.W. (1983), "Schemata in Cognitive Anthropology," Annual Review of Anthropology, 12: 429-462.

Douglas, M. & B. Isherwood (1979), The World of Goods. Towards an Anthropology of Consumption, New York: Basic Books.

Dyer, R. (1993), The Matter of Images: Essays on Representations, London: Routledge.

Frat, A. Fuat (1995) "Consumer Culture or Culture Consumed?", J.A. Costa & G. Bamossy (eds.): Marketing in a Multicultural World, Thousand Oaks: Sage, 105-125.

Ger, G. (1991), "Country Image: Perceptions, Attitudes, Associations, and Their Relationships to Context", R.R. Dholakia and K.C. Bothra (eds.): Proceedings of the Third International Conference on Marketing an Development, , New Delhi, India, 390-398.

Graburn, N. (1987), "Material Symbols in Japanese Domestic Tourism", D.W. Ingersol Jr. & G. Bronitsky (eds.): Mirror and Metaphor: Material and Social Constructions of Reality, Landham, MD: University Press of America, 17-28.

Graby, F. (1993), "Countries as Corporate Entities in International Markets," N. Papadopoulos & L. Heslop (eds.): Product-Country-Images. Impact and Role in International Marketing. New York: International Business Press, 257-283.

Hamilton, D.L. and T.K. Trolier (1986), "Stereotypes and Stereotyping: An Overview of the Cognitive Approach," J.F. Dovidio and S.L. Gaertner (eds.), Prejudice, Discrimination, and Racism, Orlando, FL: Academic Press, 127-158.

Han, C. Min (1989), "Country ImageBHalo or Summary Construct", Journal of Marketing Research, 26(2), 222-229.

Kotler, P., D.H. Haider and I. Rein (1993), Marketing Places: Attracting Investment, Industry, and Tourism to Cities, States, and Nations, New York: The Free Press.

Mackie, D.M. & D.L. Hamilton (eds.) (1993), Affect, Cognition, and Stereotyping: Interactive Processes in Group Perception, San Diego: Academic Press, Inc.

Maheswaran, D. (1994), "Country of Origin as a Stereotype: Effects of Consumer Expertise and Attribute Strength on Product Evaluations," Journal of Consumer Research, 21 (2), 354-365.

Markschark, M., C.L. Richman, J.C. Yuille & R. Reed Hunt (1987), "The Role of Imagery in Memory: On Shared and Distinctive Information," Psychological Bulletin, 102 (1), 28-41.

McCracken, G. (1988), Culture and Consumption: New Approaches to the Symbolic Character of Consumer Goods and Activities, Bloomington, IN: Indiana University Press.

Medin, D. (1989), "Concepts and Conceptual Structure," American Psychologist, 44(12), 1469-81.

Mick, D.G. (1986), "Consumer Research and Semiotics: Exploring the Morphology of Signs, Symbols, and Significance", Journal of Consumer Research, 13 (September), 196-213.

Mick, D.G. (1988), "Schema-theoretics and Semiotics: Toward more Holistic, Programmatic Research on Marketing Communication", Semiotica, 70, 1/2, 1-26.

Mick, D.G. & C. Buhl (1992), "A Meaning-based Model of Advertising Experience", Journal of Consumer Research, vol. 19, (3), 317-338.

Morello, G. (1993), "International Product Competitiveness and the #Made in’ Concept," N. Papadopoulos & L. Heslop (eds.): Product-Country-Images. Impact and Role in International Marketing. New York: International Business Press, 285-309.

Morin, E. (1984), "Pour une thTorie de la nation", Sociologie, Paris: Fayard, 128-138.

Niss, H. (1994), Made in Denmark. Nationalitetens betydning i international markedsf°ring. Aalborg University: Center for sprog og interkulturelle studier.

N÷th, W. (1990), Handbook of Semiotics, Bloomington: Indiana University Press.

Oakes, P.J., S.A.Haslam & J.C. Turner (1994), Stereotyping and Social Reality, Oxford: Blackwell.

Ozsomer, A. & S. Tamer Cavusgil (1991), "Country-of Origin Effects on Product Evaluations: A Sequel to Bilkey and Nes Review," Enhancing Knowledge Development in Marketing, eds. Mary C. Gilly, et al. Chicago: American Marketing Association, 269-277.

Papadopoulos, N. (1993), "What Product-Country Images Are and Are Not", N. Papadopoulos & L. Heslop (eds.): Product-Country-Images. Impact and Role in International Marketing. New York: International Business Press, 3-35.

Piaget, J. & B. Inhelder (1971), Mental Imagery in the Child, London: Routledge & Kegan.

Pisharodi, R.M. & R. Parameswaran (1992), "Confirmatory Factor Analysis of a Country-of-Origin Scale: Initial Results", Jhn F. Sherry, Jr. & Brian Sternthal (eds.), Advances in Consumer Research, vol. XIX, Provo, UT: Association for Consumer Research, 706-714.

Poiesz, T.B.C. (1989), "The Image Concept: Its Place in Consumer Psychology," Journal of Economic Psychology, 10, 457-472.

Schweiger, G., G. HSubl & G. Friederes (1995), "Consumers’ Evaluations of Products Labelled #Made in Europe’," Marketing and Research Today, (February), 25-34.

Seiter, E. (1986), "Stereotypes and the Media: A Re-evaluation," Journal of Communications, 36, 2.

Sherry Jr., J. (1984), "Some Implications of Consumer Oral Tradition for Reactive Marketing", T. Kinnear (ed.): Advances in Consumer Research, vol. 11, Ann Arbor, MI: Association for Consumer Research, 741-47.

Shimp, T. A., S. Samiee, & T.J. Madden (1993), "Countries and Their Products: A Cognitive Structure Perspective," Journal of the Academy of Marketing Science, 21 (4), 323-330.

Stangor, C. (1989), "Strength of Expectancies and Memory for Social Information: What we Remember Depends on How much We Know," Journal of Experimental Social Psychology, 25, 18-35

Stephan, W.G. & C.W. Stephan (1993), "Cognition and Affect in Stereotyping: Parallel Interactive Networks," D.M. Mackie and D.L. Hamilton (eds.): Affect, Cognition, and Stereotyping: Interactive Processes in Group Perception, San Diego: Academic Press, Inc., 112-136.

Stephan, W.G., C.W. Stephan, T. Stefanenko, V. Ageyev, M. Abalakina, & L. Coates-Shrider (1993), "Measuring Stereotypes: A Comparison of Methods Using Russian and American Samples," Social Psychology Quarterly, 56 (1), 54-64.

Stern, B.B. (1989), "Literary Criticism and Consumer Research: Overview and Illustrative Analysis", Journal of Consumer Research, 16 (3), 322-334.

Stroebe, W. & C.A. Insko (1989), "Stereotype, Prejudice, and Discrimination: Changing Conceptions in Theory and Research", D. Bar-Tal, C.F. Graumann, A.W. Kruglanski & W. Stroebe (eds.), Stereotyping and Prejudice: Changing Conceptions, New York: Springer, 3-36.

Volkan, V.D. (1988), The Need to Have Enemies and Allies: From Clinical Practice to International Relationships, Northvale, NJ: Jason Aronson Inc.

----------------------------------------