|
Scandinavian Political Studies, Bind 3 (1968)AMERICAN AND FINNISH JOURNALISTS LOOK AT WORLD LEADERS*Kaarle Nordenstreng Side 167
University of
Tampere and The Finnish Broadcasting Company IntroductionThe images of foreign countries and their leaders are crucial in international communication. It is evident that in the communication not only of values but also of facts (daily news), international understanding depends heavily upon the perceptions (images, stereotypes) that nations have of each other. Also the mechanism of international conflict implies the idea of reciprocal perception: besides knowing how the parties in conflict behave it is important to know how they perceive each other's behavior, before a thorough description and understanding of the conflict can be achieved.l When studying images it is not usually wise to take them individually, in isolation from other images. Cognitive perceptions always depend on relations between a (large) set of perceptions: an individual perception is to a great extent defined by its relations to other (relevant) perceptions. We perceive a person in the terms (qualities and distinctions) we usually apply in identifying and comparing people. And we look at a nation using the criteria we usually apply in discriminating between nations. Consequently, the world around us is essentially a set of interrelations. And each of us has his own way of thinking about these interrelations: each of us has a personal point of view or perspective to the world around us. Especially clear are the differences in perspectives between nations; for historical, political and geographical reasons each has its particular point of view to the rest of the world. It is not necessary to a successful communication that the parties share the same point of view, but it is necessary that they are aware of the possible differences in their perspectives. Therefore all cross-cultural studies on human perceptions2 are of great importance to the success of international communication and the process of reducing international tension. * This study was initiated when the author was research fellow at Southern Illinois University in fall 1966, and the American material was gathered during the author's tour of USA carried out with support fromtheFinnish Broadcasting Company in spring 1967. The author is grateful to Drs. James Lemert, Bruce Rucker and Erkki Teikari for their help and criticism in the planning of this study. Special gradtude is addressed to the 34 foreign editors who were kind enough to give their busy time for this study. * This study was initiated when the author was research fellow at Southern Illinois University in fall 1966, and the American material was gathered during the author's tour of USA carried out with support fromtheFinnish Broadcasting Company in spring 1967. The author is grateful to Drs. James Lemert, Bruce Rucker and Erkki Teikari for their help and criticism in the planning of this study. Special gradtude is addressed to the 34 foreign editors who were kind enough to give their busy time for this study. Side 168
Journalists are supposed to be objective reporters of the affairs of the world. However, a journalist is influenced by his own society, because as a communicator he must encode his messages in terms of his particular audience, their political interests and geopolitical frame of reference.3 It can be expected that also journalists, even experts in foreign affairs, share the general point of view of their own nations. On the other hand, it can be equally well expected that those who are daily very well informed about world affairs should have a "universal perspective" without any distortions caused by the "national perspective". The present study was designed to test the hypothesis that a national point of view influences the international frame of reference of a journalist. Seventeen American and seventeen Finnish foreign news editors served as respondents and the names of ten world political leaders were chosen to represent the international frame of reference. The choice of the world political leaders involved in this study was deliberate (and typically European): Castro, De Gaulle, Johnson, Kekkonen, Kosygin, Mao, Nasser, Thant, Tito and Wilson. The selection is by no means representative of the whole of the world, and not even of the most "important" world figures (Africa, Middle-East and Latin America are missing). On the other hand, Kekkonen and Thant do not nicely fit into the company of the other eight leaders - Kekkonen because of his minor importance in world politics, and Thant because of his special role as a "universal figure". They were included in this particular study because they were supposed to introduce some interesting aspects just on those deviating grounds. This study does not aim to draw "perceptual maps of the world" as Robinson and Hefner have done using the names of 17 countries to represent the international frame of reference.4 The purpose of the present study is firstly to ascertain whether American and Finnish journalists have the same or different points of view about the given world figures, and secondly to determine and describe in more detail the different points of view or ways of thinking to be discerned in relation to the given world figures. Given some other set of world political leaders, the results could be different. So the present study should be seen just as a limited experiment or a methodological example rather than an attempt to point out definite characteristics in the relations between world figures in general. RespondentsForeign news editors are not so common in USA as in Finland — just as the coverage of foreign material in American newspapers is relatively small compared to the share of foreign material in Finnish newspapers.5 Out of 57 Finnish dailies, 15 have their own specialists (on the average 3 in each) to cover the foreign news material. In USA only the biggest metropolitan papers (some 20 altogether6) have their own foreign news editors; usually it is the wire editor together with the city editor who takes care of foreign as well as domestic material. All national news agencies and broadcasting companies, for their part, have foreign news specialists of their own, both in USA and Finland. The background of
an American foreign news editor is very similar to that
of Side 169
his Finnish colleague. Both have considerable experience in practical journalism, and both have been travelling a good deal around the world. Also the jobs of foreign news editors are highly comparable in USA and Finland: both work in the newsroom, in an atmosphere of special importance and appreciation, and both can also participate commenting and evaluating the news they handle, although few of them have direct influence on editorial policy. Thirteen of the American foreign news editors who were used as respondents in this study were working in 8 leading metropolitan newspapers (New York, Washington D. G., Boston, Chicago, St. Louis and Los Angeles). Two of the American respondents were in broadcasting (located in Washington D. C.), and other two worked for news agencies (in Washington D. C.). All of them were men. Eleven of the Finnish foreign news editors were working for 8 national newspapers (10 in Helsinki and 1 in Tampere); 7 of these dailies represent a political party (2 conservative, 1 Swedish-speaking minority, 1 agrarian center, 2 social democrat and 1 communist). Four of the Finnish respondents came from the foreign section of the Finnish Broadcasting Company (Yleisradio) and 2 from the foreign department of the Finnish News Bureau (both in Helsinki). Three of the Finnish respondents were women. How representative are the two samples of respondents for the rest of the foreign news editors in USA and in Finland? At least the Finnish sample is statistically fairly representative, because the author knew the list of all Finnish foreign news editors and could carefully control the background characteristics in selecting the respondents (and there were no refusals). Not so with the American respondents: the contacts were more or less casual and. all who were willing to cooperate were accepted as respondents. On the other hand, there were only two refusals (both for reasons of principle: "We are for work, not for studies ..." "It is impossible and dangerous to put these things into figures ..."). And furthermore, quite a lot (maybe most) of the newsrooms were included where foreign news editors can be found in USA. So even if the American sample was not statistically very correctly compiled, it should be at least symptomatic also for the rest of American foreign news editors. The American respondents participated in the study in March 1967 and the Finnish respondents during the next two months.7 The author himself both asked the respondents to cooperate (by telephone) and served as experimenter in all the judging sessions. The respondents had reserved one hour for the judging procedure, and the sessions usually took place in newsrooms before or after a work term. Measuring TechniqueMostly it is only the multidimensional techniques that are appropriate for studying such complex phenomena as images, points of view and ways of thinking. The present study was made using a modification of the multidimensional scaling technique called similarity analysis.8 It has been designed to reveal the judgement criteria or basic dimensions that underlie the perception of a set of objects and their relations. Side 170
After a
respondent had been informed of the purpose of the
study, the experimenter "I have here the
names of ten internationally known world figures printed
on little carbon Your task is to give a rating for each pair which indicates your opinion about similarity of the figures in question. Take for instance the pair, Kosygin-Nasser. If you feel they are completely similar, that they are very close to each other in your mind, you place the two squares close to each other (experimenter shows); if you feel they are completely dissimilar, that the images are extremely far from each other in your mind, you place the squares at maximal distance (experimenter shows). Feel free to use the whole scale between the two anchor points, maximal similarity which you can think and maximal distance which you can imagine in relation to international world figures. Of course, you can judge these figures according to several different aspects, base your similarity rating on different criteria, such as the geopolitical location of the countries that the persons stand for, political views or personal style of the persons, and so on. You may ask which aspects are supposed to serve here as yardsticks for the similarity ratings, and I answer: please, don't pay particular attention to any single aspect separately and begin to rationalize — what I want is your first impression or spontaneous reaction about the similarity of the images, which is a kind of overall impression, based naturally on all of these different aspects. I don't care about sophisticated analysis of today's world situation, I ask you to tell me what you honestly think about the deeper tendencies behind everyday politics. Please, don't
feel inhibited to express your spontaneous personal
opinions. Your reactions One of the two name-squares to be compared (a "constant") was always fixed by the experimenter to one end of a 20 cm long setting panel and the other namesquare (a "variable") was given to the respondent to be placed somewhere on the continuum. The setting was registered by the experimenter by reading scale values (ranging between O = maximal distance and 100 = no distance) behind the setting panel. The order of presenting the name combinations had been randomized so that the same name never appeared in two consecutive pairs. The instruction was easily understood by the respondents, and most of them could perform the task without trouble; evidently they have been accustomed to such a comparative scanning of world figures. However, a couple of them had severe difficulties in defining their similarity conceptions, because they found each of the world figures so unique that hardly any comparison was possible. While making the settings the respondents usually spontaneously gave verbal explanations, which were written down by the experimenter. On the other hand, some respondents honestly said that the task is easy as long as they are not asked why they did so and so; they simply felt unable to explain all the factors that dictated each judgement. After giving all the 45 judgements a respondent was asked to tell the aspects to which he paid attention in making the judgements: "What do you think were the criteria you based your similarity ratings on?" This verbalized description of the way of thinking of each respondent was written down to help the interpretation of the scaling results. The 45 similarity ratings of each respondent (plus the value 100 in the diagonal to denote identity) form a similarity matrix, which indicates how similar the respondent has judged each world leader to every other world leader (cf. Figures 1-4, pp. 178-182). The data of the present study are composed of 34 such similarity matrices, 17 from the American and 17 from the Finnish foreign news editors. Side 171
Given the 34 similarity matrices, the task of data elaboration is to compare statistically these matrices in order to indicate the possible differences between the similarity rating tendencies (i.e., points of view) of individual respondents and especially between American and Finnish respondent groups. A Priori Grouping: American vs. Finnish RespondentsThe American and Finnish points of view were compared by looking how the mean similarity ratings of the American respondents differ from corresponding mean ratings of the Finnish respondents. An average similarity matrix was computed9 both for American and Finnish data. The mean similarity ratings of American and Finnish respondent groups are presented in Table 1, together with standard deviations of the individual ratings and t-values of the degree of difference. Only few significant differences can be noted between the American and Finnish ratings, when the respondents are taken as national groups. The most significant differences come out when Kekkonen is compared to Castro, Nasser and De Gaulle: the Finnish respondents see more similarity in these relations than do the American respondents. The Finnish respondents also place their president closer to Tito and Mao than the American colleagues do. The rest of the significant differences found are between Nasser and Thant, Mao and Nasser, Tito and Thant, Kosygin and Thant, and Tito and Castro; in all of these relations the Finnish respondents have seen some more similarity than the American respondents have done. It is interesting to note that comparing Johnson with the rest of the figures did not bring out very significant differences, whereas striking differences were found when comparing Kekkonen with the others. This is at least partly explained by the rather vague image and poor information the American foreign news editors proved to have of Kekkonen. Another interesting aspect is that the American respondent group rates Johnson slightly more similar to De Gaulle, Tito, Mao and Kosygin than the Finnish respondent group does. It seems to be a tendency that a respondent group sees its "own" leader as more similar to other leaders than does the other respondent group. Both of the groups have the highest similarity ratings between Tito and Nasser, Nasser and Castro, Kosygin and Tito, Johnson and Wilson, Mao and Castro, and Kekkonen and Wilson. The lowest similarity ratings, on the other hand, can be noted between Johnson and Mao, Wilson and Mao, and Johnson and Castro. The standard deviations indicate the uniformity of the similarity ratings. Both of the groups have been rather unanimous about Kekkonen's relations to Tito, Wilson and Kosygin, and about Kosygin's relation to Tito. On the other hand, neither the American nor the Finnish respondents have agreed about Johnson's relations to De Gaulle and Castro, about Castro's relations to Kosygin and Thant, and about Wilson's relations to De Gaulle. It can be
concluded that comparing the American and the Finnish
respondent Side 172
similar points
of view. Few differences can be noted in individual
cases, which reflect Empirical Grouping: Q-techniqueClassifying the
respondents into two categories according to their
national background Side 173
groups the respondents form, independent of their nationalities and only based on their way of making similarity ratings. In order to determine the mutual congruence of the similarity ratings of two respondents, a product-moment correlation coefficient was computed between them over the 45 similarity ratings. Accordingly, the similarity matrix of each respondent was correlated with the similarity matrix of eveiy other respondent. It is to be noted that the absolute level of similarity ratings was slightly different for different respondents in consequence of a personal style of using the rating scale; these level differences do not, however, influence the correlations which are only based on relative or order differences. The outcome was a correlation matrix between all 34 respondents. The correlation coefficients vary between .86 and —.18 (both between two Finnish respondents). Most of the correlations are positive, which indicates that there is a common tendency in the ratings. Both high correlations and zero correlations can be found between the two national respondent groups as well as within them, which indicates that the rating tendency really does not directly depend on the national background of the respondent. In order to reveal the potential subgroups or types within the selection of respondents, the correlation matrix was subjected to factor analysis. Thus the 34 respondents were taken as variables, which means "factoring of people" according to the idea of Stephenson's Q-technique.10 Principal axis factor analysis yielded factors with the following explanatory power: The six factors together explain 64 % of the variance caused by the similarity ratings across all the respondents. In other words, a little more than one third of the points of view or ways of thinking among the foreign news editors are unique and highly individual; almost two thirds of the similarity rating tendencies are more or less common among the respondents, i.e., shared by several foreign news editors. The communalities of individual "person variables" (cf. column h2h2 in Table 2, p. 174) indicate that there are considerable differences between individuals in this amount of common rating tendency: 4 American and 2 Finnish respondents share only about one third of their rating tendencies with the others, whereas 2 American and 5 Finnish respondents share as much as about 80 % of their rating tendencies with the others. The factorizing itself did not lead to any clear grouping of the respondents. Most of the variables had a high loading on the first factor, and the rest of the factors introduced more or less "side loadings" only. This fact can be interpreted as an indication of only one major common rating tendency among the respondents — a general "foreign news editor point of view", which is shared equally by the American and the Finnish respondents. To gain better
insight into the common rating tendencies and grouping
of the Side 174
varimax-rotation and oblique cosine-rotation. No more than 4 factors were taken into the final rotation, because beyond that factors tended to correlate heavily and also the interpretation of further factors became obscure. The varimax and cosine rotations led to somewhat different results, which is an indication of a rather complex structure of the factor space. Although some common rating tendencies could be found, the respondents share several of them simultaneously and each in his own way, so that no clear groups of respondents can be identified according to common points of view, whatever the way of looking at the factor space. Table 2 presents the varimax-loadings of the four factors. An attempt to interpret the factors will be made in the next section, where the different points of view are described in more detail. Here it is sufficient to note that both the American and the Finnish respondents have considerable loadings in each of the four factors, which means that both respondents groups have shared the same common rating tendencies. Some minor influence of national background can be seen in the fact that the first factor is mainly loaded by the Finnish respondents whereas the third factor is mainly loaded by the American respondents. This Side 175
observation points to the interpretation of the first as a "typically Finnish point of view" and the third factor as a "typically American point of view". The author hesitates to give too much weight to this dominance of these two factors by the national respondent groups, however tempting it would be for the interpretation, because the same dominance was not so clearly present in the solution of oblique rotation; it is probably a chance feature in rotation rather than a systematic tendency. Anway, the overall conclusion is clear: an empirical grouping of the respondents did not lead to a group formation according to national background. Furthermore, it is evident that no other simple grouping of the respondents is possible, either. Not even the political background of the Finnish newspaper foreign news editors seemed to influence the similarity ratings according to the political orientation towards right or left. This was already evident in the light of intercorrelations between the similarity matrices of the respondents. For instance the similarity matrix of a respondent from a communist newspaper correlated higher with the similarity matrices of respondents from a conservative newspaper (r = .51) and an American newspaper (r = .59) than with the similarity matrices of the respondents from social democratic newspapers (r between .25 and .36). It could also be expected that the medium which a respondent serves would influence his perspective and consequently similarity ratings (so that, for instance, the news agency foreign news editors would have separated from the newspapermen)11; however, this grouping hypothesis got no evidence whatsoever from the present data. Most of the variables have a higher loading on more than one factor, which means that the respondents are mosly "mixture types" who base their similarity ratings on several common points of view simultaneously — a conclusion that was already clear from the failure to rotate the factors as "pure" as possible. It is impossible to group the American and the Finnish foreign news editors of this study into any clear categories according to their modes of perceiving the relations between ten given world political leaders. But it is possible to determine the different ways of looking at these figures, which they all more or less share. The Different Points of View in Perceiving the World Political LeadersThe interpretation of the Q-factors was based both on the kind of actual similarity ratings given by the respondents and on their own verbal description of the judgement criteria. The factors reveal the basic characteristics to which the respondents have paid attention in judging the similarity relations between the world figures in question. The first factor clearly represents ideology as opposed to personal characteristics. The leaders have been judged according to their relations to political, economic and social ideologies and international power formations; typical aspects are communismcapitalism (and different variations of communism), high development-underdevelopment and neutrality—alliance. This point of view does not look at the leaders as individuals but as representatives of the larger political systems they belong to. According to this criterium, for instance, Kosygin and Tito are seen to be very far from Johnson, whereas Kosygin, Mao and Castro are seen to be quite close to each Side 176
other. The
ideological systems are obviously regarded as quite
stabile, as a kind of It should be noted that hardly any of the 34 respondents have based their similarity ratings on this ideological criterium alone. Usually it is only one point of view among others. With the reservations pointed out earlier it can be said that this ideological perspective is more common among the Finnish than among the American respondents. The second factor was given the name "National interests". Also "national policy" would fit. According to this criterium the leaders are seen more through the longrange aspirations and goals of the countries they represent than in the light of any ideological blocks or personality characteristics. The leaders are taken as if they pursued the best interests of their own country. Thus this factor represents a kind of an "emphatic point of view". It stresses the role that the leader has in the service of his country — his style, far-sightedness and effectiveness to pursue the national interests. Accordingly, this aspect brings for instance Johnson rather close to Kosygin and Kekkonen, because these countries can be seen to share much of the same interests in the long run. On the other hand, Kosygin is perceived from this perspective to be very far from Mao and Castro, whose countries have peculiar interests of their own. Also De Gaulle is seen to be quite far from Johnson. From this national interest point of view no two leaders are completely similar, because there are always some conflicts of interest between different countries, even between friends in the same power block. The lack of similarity may be also due to lack of contacts between the countries (which manifest common interests). The similarity of the national interests of two leaders is judged by simultaneously weighting the common national interests, conflicting national interests and the amount of contacts between the nations. The national
interest perspective is equally common among the
American and the The third and fourth factor introduce two slightly different points of view of the personal characteristics of the leaders mentioned in this study. If the rotation is made with only three factors, these two factors "melt together" while the first two remain the same as in the four-factor solution. This indicates that the third and fourth factor are differentiations of a more general point of view that represents the human or personality side of the leaders. While the second factor points out how the leaders act in the service of their countries, the third factor points out how they act as human beings. According to the criterium suggested by the third factor the leaders are no more "emphatically" seen through long-range national interests (not to speak about ideologies), but more like tacticians who face national as well as international problems. The similarity judgement is based on matching the practical ways in which two leaders react to situations they encounter. This point of view was given the name "Operation style". One essential aspect in the style of operation is how pragmatic or dogmatic a leader is in everyday politics. Mostly this operative point of view was obviously taken into Side 177
account by
(unconsiously) speculating what leaders in question
would do in each According to the style of operation Johnson, for example, is taken to be quite close to Mao and Castro, because if one disregards their different politics they can be seen as similar in their imaginative nationality and "tendency to ignore what people tell them", as one American respondent put it. Also De Gaulle is seen to be more similar to Mao and Johnson according to this operative criterium than according to national or ideological criteria. From this personal point of view Kosygin, on his part, is perceived to be far from Castro and Nasser, who would be relatively close to him ideologically. If the third factor represents a human aspect of the leaders in their daily operation, the fourth factor finally represents a perspective where the men are purely taken as individual personalities without any reference to international ideologies, national interests or even tactics in practical operation. The last common point of view that could be identified in the present study was given the name "Personality traits", because this criterium is based on psychological characteristics of the leaders, like strength of the personal ego, aggressiveness, emotionalism, spontaneity, moderation, quietness, flexibility, consistency, sincerety and responsibility. This personality criterium also includes the feelings of sympathy and antipathy that some of the respondents had toward certain leaders. In terms of personality traits, Johnson can be seen as rather close to Nasser and not too far from Castro and Mao, but not so close to Tito as was the case when the style of operation was regarded. De Gaulle is seen as a personality more similar to Wilson and Tito than he is seen as a political manipulator. Typical of the difference between the third and fourth factor is also that Castro and Thant are seen to be more similar regarding operation style than regarding personality traits. To put it simply, the third factor represents "pragmatism", the fourth factor "emotionalism". Both of the personality points of view, but especially "Operation style", seem to be more common among the American than among the Finnish respondents - this conclusion is once again made with the reservations presented above. However, it can be taken as a suggestion of the present study that there is a tendency among American foreign news editors to emphasize the personal characteristics of a world leader, while the Finnish foreign editors have a tendency to think more in the terms of ideological power blocks. By and large, the present factors - "Ideological blocks", "National interests", "Operation style" and "Personality traits" — can clearly be seen to represent successive points in a criteria continuum from, highly abstract international formations to very concrete personal characteristics. The whole of this spectrum plays a role in matching the world political leaders, but each foreign news editor has his own individual way of weighting the different points of the criteria continuum. In spite of the fact that no individual respondents can be taken as self evident and "pure" representatives for the different Q-factors (cf. ambiguity of the rotation), four original similarity matrices are presented below as examples of cases which mainly Side 178
fall in the four
categories revealed in this study. The examples should
not be taken Examples of Some Individual Points of ViewAs an example of using mainly the ideological criterium (Q-factor I) the similarity matrix of respondent 20 is presented in Figure 1, together with a graphical illustration of the similarity relations and factor analysis result of factorizing the similarity matrix (taken as a correlation matrix). A two-dimensional illustration of the distances between the world figures is unsatisfactory, because the similarity matrix has more dimensions as indicated by factor analysis. However, it can be used as a rough estimate to give a concrete overall picture of the structure of the similarity space. The similarity judgement criteria of an individual respondent were revealed here using factor analysis to mathematically determine the dimensions of the similarity space and the position of each world leader within it. This means grouping of the images of the ten world leaders according to the similarity ratings: each factor re Side 179
presents a group of leaders who in the mind of one particular respondent share a characteristic which the other leaders do not have. In this case, the communality values (h2) indicate to what extent each world leader is perceived to share common characteristics with the other leaders or to remain unique, i. e. within or outside the space of common judgement criteria. All of the world leader images of respondent 20 seem to fall evenly within the common similarity space, wherein four different dimensions or judgement criteria could be discovered (additional factors were residuals only, as was the case also with the three factor analyses below). It is to be noted that factor analysis is not a very suitable technique for revealing the dimensions of a distance matrix which may introduce a space with indefinite metrics (a non-Euclidean space).12 The present similarity matrix proved to have two negative latent roots, but they are so minimal (together 3 % of the sum of all 10 Eigenvalues) that they can only have a marginal influence on the four-factor results given here, and therefore they have been overlooked. Factor analysis is used here as an objective - even if a rough - means of grouping the images of the world leaders within a similarity space. The illustration shows how this mainly ideologically oriented respondent perceives Mao, Kosygin, Castro, Nasser and Tito as one "compact" communist group. Two other clusters are formed by De Gaulle together with Kekkonen and Johnson together with Wilson; they are deviations from the communist block but to different degrees and directions. Thant is perceived to be rather closely associated with some communist leaders. The factor analysis result confirms the basic group formation indicated by the illustration, but it calls forth an additional dimension which takes Thant, Nasser, Kekkonen and Tito together. This aspect can be interpreted as "small (opportunistic) mediators between East and West". As an example of using mainly the national interest point of view (Q-factor II) the results of respondent 5 are presented in Figure 2. His similarity matrix has three minor negative latent roots (altogether 7 % of the sum of Eigenvalues), which points to a slightly less consistent way of judging the similarity relations than that of respondent 20. Here the group formation in the illustration is not so clear as in the previous case. Johnson, Kekkonen, Thant, Wilson, Kosygin and Tito seem to form a rather coherent unit to which also De Gaulle belongs as a "side member" through close contacts to Kosygin and Tito. Mao, Nasser and Castro sharply deviate from this main group and each in their own directions; Mao comes closest to Thant, Nasser closest to Tito, and Castro closest to Nasser. The factor analysis result gives more insight into this somewhat unstructured group formation. De Gaulle gives rise to a nationalist interest dimension of his own ("politique francaise") in which also Kosygin, Tito and Thant have something in common - besides their other interest perspective which they share with Johnson, Kekkonen, Wilson and Thant. Nasser and Castro, on their part, form a dimension of their own, which obviously represents both specific national interest and personal leader role. Also Mao is perceived from a specific national interest point of view, Side 180
which is to some extent shared by Castro and Thant. It can be interpreted as "Asian communism interest" colored by the "messianic ethos" of the personalities in question. However, the fourth factor does not capture but a part of the variance caused by Mao, who with his relatively low communality seems to remain half unique in the similarity space of respondent 5. As an example of
using mainly the operative point of view (Q-factor III)
the The illustration shows Kosygin and Wilson on the one hand, and Mao and Castro on the other, as the two extremities between which the rest of the leaders have been placed. The first two groups given by factor analysis represent the left side of the illustration, where the pragmatists are located. The first dimension characterized by Kosygin and Wilson could be called "cool pragmatists", and the second dimension characterized by Johnson and Kekkonen could be called "hot pragmatists". The third dimension represents the lower right part of the illustration, Side 181
where the "original, unique and powerful idealists" Mao and De Gaulle are found. Also Tito belongs to this dimension, but it is to be noted that he shares the pragmatist characteristics as well. The fourth group is formed by Thant, Castro and Nasser, who might be called "tactical idealists". As an example of
using the personality criterium (Q-factor IV) the
results of Thant, Wilson, Kekkonen and Kosygin, all of whom are perceived as technocrats, form a group of "moderate rationalists". At the other extremity lies Mao with his "emotional and purposeful idealism". Factor analysis reveals that Tito and Kosygin have also something in common with this aspect of "Maoism", although it is not their main color. De Gaulle, Nasser, Tito and Johnson can be separated from the others as a group of purposeful "nationalist-egoists". Of the first-mentioned technocrats Side 182
Wilson and Kekkonen are perceived to share also some of this aspect of egopower. Like Mao, Castro has quality of his own, which is also shared by Nasser. Both of them have been perceived by respondent 22 as unpleasant and not purposeful "opportunistic emotionals" (although Nasser has another "face" in his nationalistic egoism). It is symptomatic that in the present case where personality traits are emphasized, Mao and Johnson are the figures that are perceived most as unique (low communalities). Summary and DiscussionSeventeen American and 17 Finnish foreign news editors of newspapers, broadcasting companies and news agencies defined with the aid of a multidimensional scaling technique how similar they perceived to be 10 world political leaders, taken in pairs. Each respondent made 45 similarity judgements (between each leader and every other leader). The average
similarity judgements of the American and the Finnish
respondent Side 183
groups were
compared with t-test. Little systematic difference
between the American The homogenity of the whole respondent group was analyzed in greater detail by computing correlations between the 34 respondents over the sets of similarity judgements and by factor analyzing this correlation matrix between the respondents (Qtechnique). Four interpretative factors emerged, which explain 58 % of the variance among the respondents. None of the factors are related to only American or Finnish respondents, but all the four represent both national groups simultaneously. Factor analysis also failed to group the respondents into any mutually homogenous groups, because the respondents proved to be highly individual in their judging tendencies: even if they do share most of their points of view with the others, each respondent has his own way of weighting the different common tendencies. The result suggests that among foreign news editors there are several (four) ways of thinking about the relations between world political leaders, but the differences are independent of the national background of the journalist. Neither American or Finnish foreign news editors have "national perspectives" of their own; rather they seem to share several possible "universal perspectives". The four common points of view revealed by factor analysis of the respondents (labelled "Ideological blocks", "National interests", "Operation style" and "Personality traits") represent successive points in a criteria continuum from highly abstract international power formations to very concrete personal characteristics. Each of these four common criteria was illustrated with the similarity judgements of a respondent who mainly represents one criterium only. Factor analysis was used to separate the images of the 10 world leaders in groups within the personal space of each of these example respondents. The dimensions of an image space defined by similarity relations between them at the same time reveal the different components of an individual "way of thinking". There is no reason to suspect the reliability of the measuring technique in the present case. All the respondents (except 2 who refused to participate in any quantitative study) were broad-minded (and kind) enough to see some sense in such a scaling procedure, and they seemed to take their task seriously, which made the author believe already while collecting the data that the responses were as honest as possible. Furthermore, the verbal descriptions of a respondent's judgement criteria were in good harmony with his scaling result, which on the one hand indicates that the respondents were well aware of their way of thinking, as revealed by a structured measurement procedure and, on the other, it speaks for a high reliability and validity of the scaling data. The same conclusion can be made of the factor analysis results of the four individual cases: the respondents in question are on the average 80-90 % consistent in using their four criteria of similarity judgement, and less than one fifth of the common variance remains unique, and only a minimal amount it totally unspecified (accounted for by negative latent roots). The overall
result of this study is that American and Finnish
foreign news editors Side 184
view.l3
Journalists who are experts in foreign affairs cannot be
grouped together into The present author is tempted to see this individuality and independence of national perspectives as promising and positive sign, rather than discouraging and negative result because it, first assures that international journalists think more in universal terms than in terms bound to their immediate environment, and secondly, ensures a diversity of perspectives of the political map of the world. Thus the potential diversity of the aspects seen by the communicators of foreign affairs when taken as a group, serves as a sound counterweight to the stereotype and schematic perspective of the world, which is usually found among the general public. Of course, the present research does not reveal anything about actual newswriting: this is strictly a communicator study without any reference to the messages that are produced for the public. It may indeed be so that when it comes to actual writing even the most individual foreign news editors let the audience and media influence (bias) their reporting, as suggested by the results of Cohen.14 What the present study seems to suggest is the delightful fact that the foreign news gatekeepers at least would be able to write about the world without distortions caused by national, political or media affiliations, if they wanted or apart from other influences could do so. Side 185
NOTES 1 A good frame of reference for studying international conflict as well as communication is offered by the "Stanford Interaction Model", see Holsti, 1964. 2 For instance, Holsti's study cited above; Buchanan and Cantril, 1964; and Charles Osgood's project, which at present includes 28 countries (Osgood, forthcoming volume). 3 The influence of the audience upon a journalist and his encoding behaviour is a very complex and interesting phenomenon in itself: it is not the real audience that is relevant here, but a journalist's image of the audience, which has often proved to be quite different from the real audience; see Tannenbaum and Greenberg, 1968, p. 365. The conceptions of the audience among American foreign news reporters and their influence on actual writing are discussed in detail in Cohen, 1963, pp. 108-112 and 248-263. 4 Robinson and Hefner, 1967 and 1968. 5 In the Finnish newspapers the coverage of foreign news was in 1961 on the average 21% of editorial space (advertisments and some routine material excluded); see Vehmas, 1964. In the American newspapers the coverage of foreign news can be estimated to be on the average some 5—10 % of editorial space, i.e. less than half of the same share in the Finnish newspapers; see Cohen, 1963, pp. 115-117; and Hardt, 1966; and Markham, 1961. 6 Cf. Cohen, 1963, p. 9. 7 The period during which the data were collected (Spring 1967) was quite normal and no such changes in world politics happened which would have dramatically influenced any of the world leader images in question. The respondents also could easily follow the instruction and overlook the daily events in order to judge the leaders according to their deeper and more stabile impressions. 8 See for instance, Nurminen, 1965; and Nordenstreng, 1968. 9 All computations for this study were made in the Computer Center of the University of Tampere using Elliott 803 equipment. 10 Stephenson, 1953. 11 Cf. 106-108. 12 Nordenstreng. 1968, p. 92. 13 The same overall conviction is shared . 69—81. 14 Cohen, 1963, pp. 106-112. REFERENCESBuchanan,
William, and Gantril, Hadley: How nations see each
other. Urbana, University Cohen, Bernard
C.: The press and foreign policy. Princeton, Princeton
University Press, Hardt, Jim A.:
"Foreign news in U.S. and English daily newspapers: A
comparison", Journalism Holsti, Ole R.:
"An adaption of the "General Inquirer" for the
systematic analysis of Markham, James
W.: "Foreign news in the United States and South
American press", Public Nordenstreng,
Kaarle: "A comparison between the semantic differential
and similarity Nurminen,
Anna-Marja: "Evaluation of a technique for measuring
subjective similarity", Osgood, Charles:
Studies on comparative psycholinguistics. Forthcoming
volume. Robinson, John P., and Hefner, Robert A.: "Multidimensional differences in public and academic perceptions of nations", Journal of Personality and Social Psychology, 7, (1967), pp. 251-259; "Perceptual maps of the world", Public Opinion Quarterly, 32, (1968), pp. 273-280. Stephenson, William:
The study of behaviour. Chicago, University of Chicago
Press, 1953. Tannenbaum, Percy
H., and Greenberg, Bradley S.: "Mass communication".
Annual Vehmas, Raino,
Foreign news in the Finnish morning papers, Turku,
Institute of Sociology, |