Figure 1: Cover photo by Götz Schrage. Cash-Flow 7, no. 4 (1990). © Cash-Flow/mh medienberatung + management e. U./Götz Schrage.

Figure 4: Illustration by Walter Grösel. Trend 21, goldener trend (1990): 210. © Wirtschaftsmagazin trend/Walter Grösel.

The Heads and the Walls. From Professional Commitment to Oppositional Attitude in Hungarian Sociology in the 1960–1970s: The Cases of András Hegedüs, István Kemény, and Iván Szelényi

Ádám Takács

Eötvös Loránd University, Budapest

In most of the state socialist countries in Eastern Europe, sociology remained a perpetual source of ideological quarrels from the beginning of the 1960s to the mid-1980s. With this context in mind, this paper offers an analysis of some of the decisive aspects of the development of Hungarian sociology from the early 1960s to the mid-1970s. In particular, the discussion focuses on three central figures, András Hegedüs (1922–99), István Kemény (1925–2008), and Iván Szelényi (1939), and their intellectual developments from committed and professional sociological work to the adoption of a deeply critical attitude towards socialist social development. An examination of the similarities in their intellectual development, especially as far as their political confrontation with the regime is concerned, offers a context for a discussion of some of the topical issues of the professional, institutional, and ideological aspects of academic work in state socialist Hungary and the ways in which genuine scholarly achievements could give rise to oppositional attitudes and social dissidence.

Keywords: Kádár era, Sociology, Social Criticism, Oppositional Attitudes, András Hegedüs, István Kemény, Iván Szelényi

In most of the state socialist countries in Eastern Europe, sociology remained a perpetual source of ideological quarrels from the beginning of the 1960s up to the mid-1980s. Even if party and state authorities often recognized the usefulness of sociology for their purposes, especially in periods when economic and social reforms were on the agenda, the sociological approach to the study of society never ceased to be regarded as a challenge to Marxist-Leninist ideology. The critical potential of sociology lay precisely in the fact that concrete and empirically grounded research devoted to the social facts of labor conditions, housing, lifestyle, healthcare, education, poverty, etc. tended to reveal the less familiar and gloomy side of the building of socialism. At the same time, sociology could also challenge communist ideology on its own level, namely by calling into question the social model which had been officially proposed under the label of “advanced socialist society”. By adopting sociological perspectives in their critical work on social conditions, social scientists started to discover and examine networks of relationships among social forms, stratifications, and developments which until then had gone largely unnoticed, as well as hierarchical relations of particular social strata existing and acting within the conditions prevailing under socialism. In doing so, they not only sought to rearticulate, if not explicitly to call into question, the Marxist conception of class system, but also to reconsider the Marxist-Leninist economic and social principles of the socialist model in the name of new strategies of social modernization. Thus, by the end of the 1960s, progressive Marxist sociologists in several Warsaw Pact countries, supported by reform-communist circles, tended to envisage themselves as the genuine mediators in the regime-society relationship. By taking as their starting point the empirical analysis of the given social reality, they were advocating a critical reappraisal of the ideological principles of the sate socialist regime itself.1

With this context in mind, I aim in this paper to provide an analysis of some of the decisive aspects of the development of Hungarian sociology from the early 1960 to the mid-1970s. In particular, I focus on three central figures: András Hegedüs (1922–99); István Kemény (1925–2008); and Iván Szelényi (1939). Their otherwise somewhat disparate intellectual trajectories from committed and professional sociological work to the adoption of a deeply critical attitude towards various elements of sate socialist social development and politics played fundamental roles in the subsequent formation of the profile of the cultural and political opposition, both within and beyond the social sciences in Hungary.2 I offer a comparative study of the work and careers of these scholars, whose critical attitudes towards the regime were acknowledge by the mid-1970s at a minimum with their dismissal from their academic jobs. What makes this subject worth studying is also the fact that, unlike other groups of scholars who played decisive roles in the newly forming democratic opposition in Hungary (for instance members of the “Lukács school” in the 1960–70s3 or the so-called “reformist economists” of the 1980s4), the sociologists in question never in fact formed a group. Rather, they had different backgrounds, different academic affiliations, and often contradicting views on the role and design of sociological research. Yet there were also striking similarities in their intellectual development, especially as far as their political confrontation with the regime is concerned. A comparative study of their careers and contributions, thus, offers a perspective from which to examine (1) the modes by which professional, institutional, and political aspects of academic work could play formative roles in the development of a social critical approach to state socialism; (2) the ways in which genuine scholarly achievements could influence the birth of oppositional attitudes and social dissidence; (3) the forms of comportment among party authorities, with regards to which the limits of political tolerance and the effectiveness of reprisals were always dependent on a certain ideological flexibility adapted to academic situations and on a network of formal and informal institutional and personal relations.

From Reformism to Revisionism: The Case of András Hegedüs

In the interviews he gave in the 1980s, András Hegedüs often described his political and scientific attitude in the period following his return from Moscow in 1958 as entirely “apologetic.”5 To be sure, after being prime minister in the last eighteen months of the Hungarian Stalinist regime marked by the dictatorship of Mátyás Rákosi, Hegedüs hardly seemed like someone who would have this attitude. In fact, as he reaffirmed in his memoirs, he was apologetic not only toward the socialist system that came in the wake of the events of 1956 in Hungary, but also toward the new political line represented by the Kádár regime itself. This apologetic attitude was certainly facilitated by the fact that, unlike most of the Rákosi regime’s political leaders, Hegedüs was neither expelled from the communist Party nor subjected to any disciplinary proceedings. There are reasons to believe that Kádár and his inner circle considered Hegedüs a possible ally in the fight against the revisionist tendencies within the Party, represented by the remaining followers of Imre Nagy. In 1961, Hegedüs was offered the position of Vice-President at the Central Statistical Office. It would be difficult to interpret this transfer as anything other than a reward for his loyal attitude toward the new regime. As a matter of fact, this attitude found clear expression in his post 1956 publications.6 Nevertheless, instead of taking the position, Hegedüs expressed his desire to devote himself to full-time scientific work and, more specifically, to sociology. His request received the full support of some prominent party members, including György Péter, the head of the Statistical Office, and Hegedüs was given the mandate to organize and lead the Sociological Research Group to be set up under the auspices of the Hungarian Academy of Sciences beginning in March 1963.

In the secondary literature based on the memoirs of many others in the field, Hegedüs’ name is inseparable from the rehabilitation and re-institutionalization of sociology in Hungary.7 In fact, the Sociological Research Group of the Academy was the first, and for quite some time the only, independent institute in socialist Hungary in which advanced research in the field of sociology could be carried out. Even more importantly, Hegedüs himself appeared to have been convinced at this point that sociology ought to be part of an “enlightenment process” the impact of which should spill over the barriers of even Marxist philosophy and ideology.8 Thus, by late 1963, Hegedüs’ intellectual position appeared fairly secure, and it seemed as if, over time, it would solidify even further. He was invited by the party leadership to take over the position as editor-in-chief of the political-cultural monthly Valóság [Reality], which, in line with an earlier decision of the Hungarian Socialist Workers’ Party (HSWP) Politburo, was to be turned into a journal with a “comprehensive and scientific profile.”9 Due to his new function, Hegedüs also became a member of the “Theoretical Working Group” of the HSWP, which functioned alongside the Central Committee. This move seemed a sign of an increasing political trust in him.10

This tendency, however, did not last long. One year later, at the “Nationwide Ideological Conference” of the HSWP, the journal Valóság was condemned for its ostentatious attitude and lack of self-criticism.11 Also, an important document entitled “Some Current Ideological Tasks of the Hungarian Socialist Workers’ Party: Guidelines by the Central Committee,” which was approved by the Central Committee in March 1965, harshly criticized sociological research in Hungary for its “abstract reasoning” and its “uncritical borrowings from the dubious achievements of bourgeois sociology.”12 The Guidelines also condemned Valóság for its “erroneous views,” “incorrect, bourgeois attitude,” “oppositional tendencies,” and “decadent approach.”13 These acts earned Hegedüs the label of “revisionist” for the first time and eventually triggered his removal from his position at the journal in June 1965.

For a time, however, Hegedüs’ dismissal from Valóság brought about no drastic change in the course of his intellectual career. On the contrary, his critical behavior had in fact channeled him towards the reformist party circles within the HSWP leadership, whose importance happened to be on the rise due their role in preparing the new economic reforms to be launched in January 1968. In 1965, Hegedüs was invited to take part in the work of the “Preparatory Committee for the Reform,” run under the auspice of the Economic Board operating next to the Central Committee. Hegedüs was asked to organize and oversee one of the eleven workgroups designed to assist the Preparatory Committee. His group was tasked with investigating “interaction between economic and social relations.”14 In this period, with respect to his own scientific work, Hegedüs continued to both extend and sharpen his theoretical sociological research. On the one hand, he devoted himself to critical analyses of socialist society from a structural point of view.15 On the other, he pushed the limits of critical analysis to new levels concerning the place and role of sociology within the system of Marxist social sciences, as well as concerning sociology’s claim to tackle some of the most vital social problems related to the building of advanced socialist society in general, and in Hungary in particular.16

Without doubt, the emphasis on sociology’s task of providing scientific self-knowledge for socialist society sums up the credo of Hegedüs’ vision of Marxist sociology. But it is also clear that his radical reformist endorsement of the critical function Marxist sociology could play in socialist society is separated, if at all, by only a thin line from the promotion of truly “revisionist” ideas. After all, Hegedüs’ Marxism seemed ready to jettison the classic Marxist theses on social development the moment the sociological analysis of the concrete social realities proved them false. At this point, however, he also believed that his ideas had essentially been confirmed by recent political and social developments and especially by the new economic reforms under preparation in both Czechoslovakia and Hungary. In an essay published in 1967 in the Hungarian literature monthly Kortárs [Contemporary] entitled “Reality and Necessity: The ‘Self-Criticism’ of Socialist Society as a Reality and a Necessity,”17 he went so far as to assert the “historical necessity” of the emergence in socialist society of a new type of critical attitude designed to reshape the relationship between the party and society.

The fact that in August 1968 it was not sociologists, but Warsaw Pact troops who readjusted the regime-society relationship in Czechoslovakia ultimately triggered the escalation of Hegedüs’ situation within his party. On August 21, along with the Hungarian sociologists and philosophers protesting in Korčula, the Party members of the Sociological Research Group condemned the intervention and Hegedüs addressed a petition to the Central Committee on the issue. In its report to the Politburo on this case, the Scientific, Educational and Cultural Board of the Central Committee made it clear that the protest issued by Hegedüs and his comrades against the intervention in Czechoslovakia was in reality only the most recent chapter in a far-reaching story. The document noted that since 1966, the Agitprop Committee had brought up the issue of the “negative tendencies” manifested in “Hegedüs’ theoretically and ideologically dubious ideas” several times. In conclusion, and as a way to solve the situation, the report proposed the removal of Hegedüs from the leadership of the Sociological Research Group.18

After 1968, Hegedüs’ research activity in sociology underwent a reorientation. In a somewhat programmatic study entitled “For the Healthy Development of Marxist Sociology,” which Hegedüs wrote right after his removal from the Sociological Research Group and which was published in the Társadalmi Szemle [Social Review], the theoretical monthly of the HSWP,19 he urged the continuation and even intensification of sociological research on social structures and social stratification under socialism. Hegedüs nevertheless did not entirely abandon the idea of advocating a radical “reformist position” when it came to sociological issues related to socialist development. This ambition, for example, was clearly manifest in the studies he devoted in this period to the sociological analysis of the question of “bureaucracy” under socialist conditions.20 Also, his views on “social progress” under socialist circumstances soon came under ideological attack.21 To be sure, it was precisely with reference to these lines of research that the accusations of revisionism against Hegedüs could be relaunched, accusations which eventually would lead to his expulsion from the party and his exclusion from academic and cultural life in general.

After all, by the end of 1972, the Kádárist party leadership had come increasingly under pressure from both its own hardliners and Moscow, each of which were demanding a revision of the allegedly overly liberal economic policies of the party.22 Under these circumstances, Kádár was all too keen to demonstrate, to those inside and outside of his party, that the reform of the Hungarian economy and society was firmly under the control of the HSWP and that no deviation from the official Marxist-Leninist dogmas would be tolerated. As a result, there was a sudden change in the ideological climate and in the line that divided what could be tolerated as a legitimate Marxist “discussion” of the questions of existing socialism and what was to be rejected on the grounds of its assumed anti-Marxist content. Not surprisingly, the ideas defended by Hegedüs, along with those promoted by the members of the Lukácsian Budapest School, fell soon prey to this ideological fervor, which sought to cleanse Hungarian Marxism of its new leftist wildings.

In January 1973, speaking before the Nationwide Ideological Conference in Budapest, György Aczél, the Agitprop Secretary of the Central Committee, left no doubt about who was to blame for “denying the existing socialist practices.” He named Hegedüs among others, and he accused him of “calling into question the fundamental theses of Marxism.”23 As a consequence, during a debate held in March 1973 under the auspices of the Cultural Political Work Collective the severely “anti-Marxist platform” of several social scientists and philosophers, including Hegedüs, Mária Márkus, Mihály Vajda, Ágnes Heller, György Márkus, György Bencze, and János Kis, was unanimously condemned.24 On the basis of this report, the Central Committee of the HSWP prepared a proposal for the Politburo. The Politburo accepted the proposal and decided to publish a final resolution on the case.25 It also ordered the Hungarian Academy of Sciences to take several measures against the scholars in question. Hegedüs, Vajda, and Kis was expelled from the HSWP, and all the scholars involved were dismissed from their academic jobs on the grounds of their “incapability for scientific work.” (They were offered lowering-rank positions as scientists or research assistants.)

Since none of the social researchers in question accepted the new jobs offered by the Academy of Sciences, their academic carriers in socialist Hungary were definitively over. As far as Hegedüs was concerned, after having accepted various advisory positions in large communist companies and after having been quickly dismissed from them at the order of party authorities, he retired in 1976.26 His sporadic collaboration with the increasingly significant democratic opposition movements in Hungary during the 1970 and 1980s had often been hindered by his unbroken belief in the possibility of a pluralistic socialist society without the implementation of a pluralistic political party system. But his role as critical sociologist and his vision of the enlightened moderation of the society-regime relationship in communist Hungary were doomed to be relegated, at least until the end the socialist period, to the realm of academic folklore. In a volume published in English on Hungarian sociology in 1978 and edited by his successors at the Sociology Institute of the Academy of Sciences, the main text of the brief introductory study devoted to an assessment of recent sociological research in Hungary made not a single mention of his name.27

The Empirical and the Illusionary: The Critical Sociology of István Kemény

One could characterize the sociological career of István Kemény as that of a strong character who was recurrently compelled to do empirical analyses of delicate topics—social stratification, poverty, the conditions of working class, the behavior of economic leaders, the problems faced by the Roma populations—related to the first two decades of the socialist reality in Hungary under Kádár. The unusual nature of his career was determined by the very historical event that served as the alpha point for both his career and the regime itself, namely the 1956 revolution. Having originally been sentenced to four years in prison for allegedly having participated in a “seditious conspiracy” during the revolutionary events, Kemény was released from prison in 1959.28 Between 1960 and 1969, he worked as librarian at the National Széchenyi Library in Budapest. In 1963, he was asked to join as assistant a newly launched group research project conducted by the Central Statistical Office on the question of “social stratification” in Hungary. By accepting this invitation, Kemény succeeded in part in adopting sociology as his main profession and became involved in one of the most instructive and challenging empirical sociological research projects in Hungary at the time.29 Since it used the term “stratification” (“rétegződés”) as one of its keywords, the 1963 survey challenged the view according to which the tendency of socialist society to lose gradually its original class structure should necessarily be understood as an improvement towards social homogeneity. In fact, as the project demonstrated, the loosening of class constraints had led to a more differentiated and not less imperious system of social stratification.30

In 1969, Kemény was asked to join the Sociological Institute of the Hungarian Academy of Sciences as a full-time research fellow. This change of status meant that Kemény immediately became involved in several empirically based research projects initiated and run by different institutes. One of the most interesting among them was devoted to the so-called “low income population” in Hungary. This research was run in effect by a work group within the Central Statistical Office. According to Kemény’s memoirs, the interest in the study of “poverty” in socialist Hungary was already present in the 1963 national survey, but György Péter, the president of the office, firmly opposed this idea, since he believed that if the Office as state institute “started to study poverty, this would suggest that it [socialist Hungary] was a system in which poor people could be found.”31

Kemény’s participation in the survey and the attention he devoted to the living conditions of the “low income” population became the foundation for his reputation. It can be said that this was one of the groundbreaking research initiatives in which he proved himself as a sociologist working with statistical means, but willing to go beyond the simply descriptive level of survey data to analysis of what these findings reveal about the living conditions under state socialism. Kemény’s task of translating the statistical category of “low income” into terms of the people’s real living conditions, and especially the living conditions of blue-collar workers, revealed hitherto unnoticed—or rather denied—aspects of socialist reality. Even before this research project officially terminated in 1972, he had an opportunity to give a lecture in 1970 on the topic at one of the annual sessions of the Hungarian Academy of Sciences, in which he did not hesitate to talk about the phenomenon of “poverty” in socialist Hungary. In fact, he was in all likelihood the first social scientist to use this term openly in an academic speech in the post-1956 period in Hungary.

In his talk, Kemény claimed that it was misleading to draw a strict limit based on a minimum income per head in a household, as was proposed by the Central Statistical Office, in order to define a person or a family as belonging to the “low income” category. Instead, he argued that the descriptive use of the “low income” category should include consideration of concrete living and housing conditions, including family composition, cost of transportation, whether someone lived in or had a sublet, whether someone lived in an urban or rural setting, etc. According to Kemény, this would enable a more nuanced understanding of the poor as people “who were not able to live like others do.”32 With this definition in mind, Kemény was keen to demonstrate new social inequalities in the socialist reality in Hungary. According to him, poverty as a real condition affected the lifestyle, social habits, educational standards, and everyday practices of those concerned.33

Not surprisingly, Kemény’s talk at the Academy created instant havoc in the Party headquarters. Although initially both Népszabadság [People’s Liberty] and Társadalmi Szemle published positive overviews of Kemény’s talk (which, however, failed to mention the term “poverty” in their account), more drastic consequences soon followed.34 Kálmán Kulcsár, the head of the Sociological Institute at the time, was immediately ordered to dismiss Kemény from the Institute. Kulcsár did as he was told, but since Kemény had already been conducting another ongoing survey in the Institute concerning the Hungarian Roma populations, the Party headquarters was contacted again in order to determine what to do. Finally, the decision was made to allow Kemény to keep his job on a monthly basis, i.e. by “signing on the first day of each month a work contract which would last to the last day of the month” and repeating this until the survey was completed.35 Kemény finished his survey on the Roma in late 1972, after which his status at the Institute was terminated.

The aim of the 1971 survey on the Roma population was to offer a comprehensive view of the social situation of Roma in Hungary, including their “linguistic and ethnic composition, settlement types, regional distribution, housing conditions, family size, number of children and live births, education, the effects of industrialization in the 1950s and 1960s, employment, and income levels.”36 Nonetheless, the research carried out under Kemény’s leadership between 1970 and 1971 was new and unusual in several respects. Most importantly, in setting up the basic analytical categories of the project, Kemény refused to attribute particular importance to the ethnic character of the population under study. As he stated, “in our research we classified as Roma all people whom the surrounding non-Roma community considered Roma.”37 The enabled him and his team to sidestep the task of providing a scholarly (chimerical) definition of who was Roma and who was not, but perhaps more importantly, it allowed them to focus their efforts on what they considered the essential sociological aspects of the population under study. “The Roma question is fundamentally not an ethnic question, but a question of social strata,” the study concluded in the summary of its findings.38 This indicated that Kemény’s sociological approach to the Roma followed in the footsteps of his earlier survey on poverty in that he privileged questions of social stratification over questions of social segregation or ethnic identity. Also, Kemény was far from sharing the optimism of some of the communist leadership, who considered the rapid transformation of the working and living conditions of the Hungarian Roma population as an unqualified form of progress towards social assimilation. Although the 1971 survey confirmed the facts related to the drastic changes in employment and to some extent the amelioration of living conditions, in other areas (especially in housing and schooling practices) it noted severe drawbacks.

If the 1971 survey on the Roma population did not cause a political scandal, this was due primarily to its accuracy and the indisputably scientific nature of its methods, but also to the fact that the circumstances of the Roma communities were far from being in the forefront of academic or social debates in Hungary at the time. Nevertheless, the whole body of the research material was released only in 1977 as an internal bulletin published by the Sociological Institute of the Academy, and it had no table of contents and no ISBN number.

By the time the Roma survey had been completed at the end of 1972, Kemény’s monthly based contract at the Sociological Institute had expired and had not been extended. Meanwhile, the historian Miklós Laczkó, who at the Institute of History of the Academy of Sciences was given the task of preparing a research project on the Hungarian working class, contacted Kemény and asked him to do a survey on Hungarian workers.39 The Institute of History itself was asked by the Scientific Board of the Central Committee to carry out this research, and the director of the Institute, Zsigmond Pál Pach, was convinced by Laczkó to employ Kemény for this task. To be sure, this choice was not unfounded, since Kemény’s earlier research on the working class had even drawn some attention in broader public forums.40 But a closer look at this situation reveals very well the inherently contradictory and unstable processes through which, in an academic environment, communist functionaries sought to assess the party’s ideological expectations. In fact, the research initiated at the History Institute on members of the working class had already clearly indicated the changing ideological circumstances which, in the short run, had brought to a standstill the economic reforms and triggered the official political rehabilitation of the doctrine of the “leading role of the working class” in Hungary. Under these circumstances, Kemény, with his 1956 legacy and bad academic reputation, did not in principle have a chance to return. But precisely because ideological and scientific expectations were suddenly and inextricably mixed, informal ties gained increasing significance. Pach was undoubtedly a loyal party functionary, but he could be convinced to take the risk of reinterpreting the meaning of “ideologically sound” as a characterization of potential colleague in light of an alleged need of expertise. And in doing so, he was clearly ignoring the fact that Kemény had already been prohibited from carrying out academic research in another scientific Institute belonging to the same establishment.

The survey on the Hungarian workers began on September 1972 and was finished by the end of 1973. In part, Kemény used most of the descriptive categories developed in his earlier research on social stratification, the working class, poverty, and the Roma populations, applying these categories to workers.41 One of the most striking aspects of Kemény’s descriptive study on the working class was the strong emphasis on the forms of social cohesion, which correlating closely with workers’ morale. Workers showed significant shared commitment to common concerns, including mutual recognition of expertise, solidarity in struggles for better earning and working conditions, and a shared interest in technological improvement. However, in light of Kemény’s survey (which was based on interviews), these forms of cohesion were delineated as forms of common strategies of negotiation and tactics of circumvention directed against the various forms of administrative power represented by the management and directors of the factory or the party. In the preface to the French edition of his book on workers, Kemény described the general strategy followed by the Hungarian working class as one of “permanent resistance,” according to which they sought “to obey the instructions in appearance only.”42

In 1973 the Scientific, Educational and Cultural Board of the Central Committee organized a debate at Institute of History on Kemény’s manuscript. The text was harshly criticized by leading Hungarian historians, such as Iván Berend T. and György Ránki.43 This was followed by a series of events which adhered to a well-known political logic. First, Kemény’s manuscript on the Hungarian working class was rejected for publication. Then, in March 1974, the Institute of History was ordered to terminate his contract, and virtually at the same time Kemény was prohibited by the Party authorities from participating in any research or publication initiatives. In 1975, the National Educational Institute led by Iván Vitányi tried unsuccessfully to hire Kemény to take part in a research project.44 After this, Kemény attempted to engage in various research initiatives using his colleagues as cover, but in January 1977, he decided the situation was hopeless and resolved to leave Hungary for France.

Iván Szelényi and the “Immanent Critique” of Socialist Society

In a recent essay written on the development of Hungarian sociology in the 1960s, Iván Szelényi argued that between 1966 and 1968, Hungarian sociologists began to realize that empirical research in itself does not necessarily lead to value-free or apologetic results. Empirically grounded sociological investigations were increasingly perceived as having the potential to provide critical insights into the social determinants of socialist society.45 According to Szelényi, by the end of 1960s, there were two general but not mutually exclusive trends that provided the impetus and the intuitive backdrop to these critical approaches. On the one hand, there was an approach which aspired to offer an “ideological critique of socialist society.” This approach was influenced by György Lukács46 and his school, and it was championed by Hegedüs. It sought to contrast the reality of established social conditions in existing socialism with the Marxist ideals. A different approach, on the other hand, was advocated by more empirically-minded sociologists, such as Szelényi himself, who were carrying out a “critique of socialist ideology.” This approach focused on some of the internal inequalities and contradictions of socialist society, which reflected the regime’s ideological blind spots and therefore favored the elaboration of an “immanent” critique of socialist ideology and social reality.47

Szelényi’s account of this topic is worthy of consideration from a historical point of view in part because in some of his writings published in the early 1970s he had already made clear his position on the critical function of sociology. In fact, in the methodological part of his dissertation Settlement System and Social Structure, submitted in 1972 for the degree of “candidate of science” (kandidátus, the equivalent of a PhD degree), he outlined the principles of social criticism in sociology in terms very similar to those presented in his more recent writings. In his dissertation, Szelényi drew a sharp distinction between “social critique” and “critique of ideology,” and he argued that, unlike the former approach, which appeals to transcendent values in order to influence collective will and prompt action allegedly needed to build a better society, the latter seeks to analyze ideology critically as a social product serving actual interests.48

Szelényi joined the Sociological Research Group of the Academy in 1963 at the invitation of Hegedüs, first as a part-time research fellow and then, from 1967, as a full-time research fellow. His first work on housing conditions in one of the slum-areas in Budapest (coauthored with Ferenc Nemes) marked his entry into the field of sociology.49 In 1968, Hegedüs was forced, for political reasons, to resign from the leadership of the Sociology Group, and his position was taken over by Kálmán Kulcsár. At the time, Szelényi was tasked with the part-time supervision of the newly established “Sociological Laboratory” at the Social Science Institute, working under the Central Committee of the Party. A year later, he published his work (coauthored with György Konrád) on the sociological problems faced by the communities living in the new housing projects in Hungary,50 and shortly after this, he also took over the direction of the regional sociological department at the Institute of Sociology of the Academy of Sciences. Simultaneously, Szelényi also began a teaching career at the Karl Marx University of Economics in Budapest, and he similarly was given a teaching position in sociology at the Political Academy of the party.51 Thus, when Szelényi was appointed to serve as one of the editors-in-chief of the newly established sociological monthly Szociológia in 1972, his career seemed to be on a fast track to ultimate recognition. As a matter of fact, at that point in time he was undoubtedly one of the highest-ranking social scientists in Hungary who was not a member of the HSWP.

To be sure, Szelényi’s success was influenced by the fact that he kept his distance from sensitive political matters. For instance, unlike many of his prominent colleagues (including Hegedüs), he refused to denounce publicly or through official Party channels the Warsaw Pact invasion of Czechoslovakia in 1968, and he also remained reserved with regards to the Lukácsian-Marxist social critical attitude widespread in the Sociology Group. As he later remarked, not only did his empirical mindset save him, for the time being, from getting into political trouble, but he also managed to benefit, in his career, from the overall intellectual and political situation.52 Nevertheless, one should note that the topics he chose and the approaches he followed in his research allowed him to move in directions that were far from any simple value-free empirical position. In fact, in a study published in 1969 on the role of sociology, Szelényi argued that the empirical orientation in sociology had the genuine potential to foster normative interpretations in social policy or open up alternatives for social services.53 

In a way, in their 1969 book Az új lakótelepek szociológiai problémái [Sociological Problems of the New Housing Developments], Konrád and Szelényi had already gone beyond a mere descriptive account of the case under study. Without doubt, some of the concrete findings of their investigations were truly shocking.54 Most notably, statistical evidence showed that, quite contrary to what was expected, apartments in newly built housing developments in Budapest and other major cities appeared to be systematically allocated to people belonging to social groups with higher incomes, mostly to the educated middle and upper middle class. On the base of these findings, Szelényi and Konrád revealed the de-privileged status of low-income earners and the working class as such and concluded that “as a whole, the construction of new housing developments cannot be characterized as social or communal house-building” in Hungary.55 Furthermore, they called attention to the “exceptionally grave consequences” which these developments were about to create in a metropolitan environment in terms of “social segregation.”56

Between 1970 and 1973, Szelényi and Konrád extended and deepened their analysis of the Hungarian housing system. In 1972, Szelényi submitted a manuscript entitled “Settlement System and Social Structure: Sociological Elements for an Analysis of the Hungarian Housing System and Urban Structure” to obtain a PhD degree.57 The text provided a more radical assessment of the problems related to the housing issue in Hungary, and it also embedded these problems in a larger socio-historical and structural analytical framework. Sociological problems concerning housing were thus found to be representative of other major forms of socio-economic inequalities under socialism, and this called for further investigations. Also, one of the novelties of the new analyses was their emphasis on the evaluative and critical importance of sociological analyses addressing the urban housing and planning system. As Szelényi stated in the methodological part of his dissertation, an immanent “ideological-critical” approach defined as “sociology of planning” was necessary in order to reveal and assess the “social relations of interest” underlying the processes of socialist social planning.58

In 1972, Társadalmi Szemle published an article, which was strongly critical in tone and in content of a paper published by Szelényi and György Konrád a few months earlier on various sociological and historical aspects of Hungarian urban development.59 The vehemence of the article was hardly surprising if one takes into account the purpose and arguments of the paper it was targeting. In a nutshell, by labeling urban development in Hungary “retarded” or “lagging,” Konrád Szelényi managed to blame the socialist economic policy of the previous two decades for its neglect of proper urban infrastructural developments, criticize its insensitively administered social-policies, and point out some current “social conflicts” which had been consequences of these wrong-headed policies.60

In one of his late interviews, Szelényi characterized this ill-received writing as the best he had ever written with Konrád.61 Whatever the case may be, it is certain that in the beginning of the 1970s, with the rise of anti-reform sentiments and the new anti-reform ideological offensive in the making, the critical approaches and orientations advocated by the Konrád and Szelényi tended to fall short of meeting the new prerequisites set forth for a “legitimate” Marxist way of doing social scientific research. Apart from the growing pressure to reinstate a noticeably more orthodox Marxist ideological approach to both theoretical and empirical issues, political approval (and disapproval) began to play an important role in shaping sociological research topics and activities. Even the ambition to exert more straightforward political control over the sociological research apparatus appeared on the agenda, as demonstrated for instance by an Agitprop party document from 1973 which proposed subjecting sociological surveys to “central authorization” in order to prevent them from being used to draw “false” or “ideologically hostile” conclusions.62

In principle, given his leading positions at various research institutions and the fact that he had been elected to serve on the editorial committee of the newly established revue Szociológia, Szelényi seemed to have little to worry about. Yet, in a way, it was precisely his personal inclination towards professional solidarity and his belief in the pursuit of sociology as an independent critical science that would soon bring him close to the end of his prosperous career in Hungary.

In 1973, Szelényi was among the few intellectuals who protested against the denunciation and removal from their academic positions of some of the closest disciples of Lukács and sociologists like Hegedüs and Maria Márkus. The next political event in which Szelényi took an important part was the trial of Miklós Haraszti. Haraszti, at this time an ultra-leftist poet and writer, was arrested in May 1973 on charges of having distributed mimeographed copies of his work entitled “Darabbér” (“Piecework”), which had not been given approval for publication. In the trial, Szelényi agreed to testify that as a journal editor, he intended to publish parts of Haraszti’s text in the revue Szociológia because he considered it a valuable and realistic analysis of factory life and workers’ lives in Hungary.63 To be sure, this statement was not entirely true. Nevertheless, due to the appropriate strategy chosen by the defense and the solidarity campaign that surrounded the case, Haraszti, although found guilty on the charges brought against him, was sentenced to serve only eight months in prison, a sentence which was suspended on condition that Haraszti spend three years on probation. But the trial had other consequences as well. Because of his involvement is the case, Szelényi was removed from the positions he held at the Institute of Social Sciences and at Szociológia. He was also temporarily prohibited from publishing, but more importantly, his reputation as a critical but reliable non-party member academic was severely damaged.

Interestingly enough, by the time they got involved in the Haraszti trial, Konrád and Szelényi had already embarked down a path to challenge the regime in power directly. At Konrád’s initiative, they had started to compile a scholarly manuscript which Szelényi envisaged as their critical-sociological masterpiece. To be sure, they were well aware from the very beginning that the task was politically impossible, meaning that the text would never be published in Hungary. As Szelényi later remarked, they were consciously preparing themselves to “commit scholarly suicide.”64 Their manuscript was thus meant from the outset to be a samizdat in its format, which makes it one of the very first examples of this genre in socialist Hungary.65 Not surprisingly, the police, which had been keeping Konrád and Szelényi under constant surveillance since the Haraszti trial, was well informed about their activities and waited for the moment to confiscate the manuscript and arrest its authors. The two men were detained on October 22, 1973 on charges of incitement, and they remained in custody for seven days.

The major argument of Konrád and Szelényi’s samizdat book The Intellectuals on the Road to Class Power was that under Eastern European state socialism, the intelligentsia was in the process, for the first time in history, of forming a dominant class. With this context in mind, the authors sought to adopt a reflexive critical position, which, like in Szelényi’s earlier sociological works, aimed to provide an immanent “critique of ideology.”66 According to Konrád and Szelényi, the class dominance of intellectuals in state socialism manifested itself in their increasingly crucial position (and allegedly experts) as “planners” and “redistributors” within this system. In other words, they argued that a constant intellectual materialization of certain “teleological” knowledge about long-term public interests in socialist society, embodied in the intellectuals’ bureaucratic class position, not only played a functional role in sustaining the regime, but was a fundamental element without which the socialist mode of production itself would have lost its distinctive features.

Without a doubt, Konrád and Szelényi’s book was a clear attempt to call into question some of the most crucial ideological cornerstones of existing socialist regimes: the tenet of the leading role of the working class and the ideological benevolence of the party. As a matter of fact, this point was clearly stressed in a report submitted to the Politburo of the HSWP about the case.67 The document also informed its readers of the outcome of this “unlawful activity”: after seven days of detention, the two suspects acknowledged authorship of the manuscript, and the case was closed with a “prosecutor’s warning.” At the same time, as a result of the case, Szelényi immediately lost his remaining jobs at the Institute of Sociology and the University of Economics, and his career in sociology and in the academic life in general was definitely over. The only reasonable option for him was to accept the at offer made by the interior affairs authorities, which at the time was rather exceptional, to leave the country.68

Conclusion: From Professional Commitment to Oppositional Attitude

The most striking aspect in the careers of István Hegedüs, István Kemény, and Iván Szelényi is not simply that, even with their different intellectual and political backgrounds, fields of interest, and academic contributions, they were all sidelined by the mid-1970s for political reasons. Even more remarkable than this is the fact that their involvements in politically contentious situations were triggered by the adoption of a similar intellectual attitude. Nevertheless, the formation of their noticeably analogous way of perceiving and reacting to certain scholarly situations seems to imply more than mere discontent with certain ideological expectations in Hungarian academia. It stemmed rather from their engagement in a complex setting of professional, institutional, and disciplinary practices and relations that gradually shaped their personal experiences and scholarly strategies in a similar way.

From this point of view, the decisive impact of two institutions (the Central Statistical Office and the Sociology Research Group) on the development of the intellectual profile of Hegedüs, Kemény and Szelényi should be highlighted. Although the forms and lengths of their engagement in the work of these institutions varied greatly, similarities are also apparent. Contact with the pioneering sociological work carried out at the Statistical Office constituted an important milestone in the career of all three of them. It certainly made them appreciate the role of a specific institutional environment and a diverse academic body in the development of an effective and relatively free research agenda. Hegedüs seemed to have been fully aware of this when he was given the green light in 1963 to establish his Research Group at the Academy of Sciences, where he also hired Szelényi. At the same time, as Kemény has remarked, the Sociology Research Group represented a trend similar to that of the more empirically minded sociological cluster of the Statistical Office led by Zsuzsa Ferge insofar as both institutions “wanted something that was hitherto forbidden” in Hungarian sociology.69

It should be noted that this took place during the subsequent period (1968–72), when the multiplication of institutions and research opportunities allowed for an increasing flexibility in sociological research and teaching. This was illustrated for instance by the case of Szelényi, who divided his time between the Sociological Research Group and the Institute of Social Science of the Central Committee, while he also held various teaching positions. The emergence of this new situation within the sociological profession in the early 1970s was certainly fostered by the central administration’s growing interest in and demand for accurate social knowledge relevant to various policy and economic issues. For example, Kemény’s research on the Roma population and the working class and Szelényi and Konrád’s work on the housing conditions in Budapest and other cities clearly reflected this tendency. This conjuncture in sociology has led to the proliferation of research institutions and even the introduction of a certain division of labor between them, and it has also created a need to implement forms of professional training to ensure further reinforcement. At the same time, this new situation has also changed the ways in which institutions in the academic sphere are used by sociologists to adopt and pursue their research agenda. Kemény’s pursuit of various research projects in different institutions between 1969 and 1973 demonstrated significant flexibility in this regard. To be sure, the growth in the available resources (including financial resources) and the reliance on project-oriented institutional backing have created significantly more options for research, much as they have also enabled people working in the discipline to pursue their efforts with a greater degree of professional commitment and have made it easier to overlook built-in ideological safety mechanisms in research.

Apart from the institutional factors, the variety of topical interests and approaches in sociological research and the ways in which the image of Hungarian society was altered over the course of the 1960s in sociological debates have clearly shown a strong vivacity and an openness within the discipline. In this context, both the more social critical approach taken by Hegedüs and the empirically driven orientation developed by Kemény and later Szelényi shared the conviction that society was made up of critically important factors which have their own particular functions and modes of development. The focus on social stratification on the one hand and the mesmerizing effects of discovering and analyzing social inequality on the other also constituted a common element in their works. Thus, Hegedüs’ strong insistence on the function of sociology as the most direct scientific instrument in the pursuit of critical knowledge of society has not essentially contradicted the more empirically grounded approaches adopted by Kemény and Szelényi. The differences between their approaches were, rather, strategic, insofar as Hegedüs insisted on the fact that the importance of sociological research should lie in ushering academic discourse towards an explicitly social critical, if not political role—something which Kemény and Szelényi were less ready to embrace if it was propagated in the name of a normative, let alone Marxist perception of society. For them, the realistic tone of sociology implied in and of itself a sufficient stance in order to approach social reality in critical terms.

This strategic difference was also reflected in the different ways in which Hegedüs, Kemény, and Szelényi appealed to and used Marxism in their works. Although they all seemed to agree fundamentally that orthodox Marxist-Leninist categories were totally inadequate for a sociological analysis of social structures and development, they nevertheless manifested different rationales in their precepts on which their rejections were based. In the case of Hegedüs, his adherence to the idea of socialism remained unbroken throughout his career. It was precisely this idea that fueled his criticism both of the Stalinist vision of society and the more technocratic agenda of building socialism. For him, redefining socialist reality in terms of domination, subordination, alienation etc., and thus challenging the received doctrines of Marxism-Leninism, was a necessary consequence of the perception of sociology as the ongoing critical examination of the course of socialist construction in which the drive towards “optimal” economic and social development should be counterweighted by a particularly strong focus on processes of “humanization.”70 Thus, for Hegedüs the constructive use of Marxism in the search for a “leftist” normative view of society remained a cornerstone of his sociological approach. Kemény’s manifest rebuff of Marxism followed a different path. In his case, it was more the result of a pragmatic rejection expressed in neutrality towards, neglect of, and cavalier disregard for Marxist categories. However, this sociologically orchestrated disinterestedness was grounded in the very methodology he employed in most of his research. The combination of social-statistical quantification with deep interviewing offered empirical findings and a foundation for social categorization which were substantial proof of the purely apologetic nature and scientific inadequacy of official Marxism-Leninism. In Szelényi’s case, the motivations for overlooking Marxism were different in nature. His stance was based on a predominantly theoretical rejection, manifested in a strategy of almost complete neglect of Marxist terminology in his earlier writings, which has accompanied an increasingly subtle search for empirical confirmation. In other words, for Szelényi, the inadequacy of the Marxist approach has relied primarily on its incapacity to address and frame phenomena of social structure, stratification, mobility, inequality, etc. in the study of which other Western sociological theories (for example the Weberian or Polányian approaches) have proven more conclusive. Nevertheless, it was precisely the fact that his rejection was theoretical in its design, and not purely political or empirical, that allowed him to return in a certain way to Marxist categorizations in his The Intellectuals on the Road Class Power.

Nevertheless, it would certainly be misleading to characterize the series of events that led to the exclusion of the three sociologists in question from Hungarian academic life by the mid-1970s as a cumulative process which could not have gone another direction and ended in a different scenario. Although there are always underlying reasons for ideological climate changes (and in this case, they usually accurately reflected the actual orientations and power struggles in which policy was rooted, especially in the academic sphere), ideology as such was far from a coherent and all-powerful system of norms providing direct support for the eventual implementation of administrative measures. Ideological intervention had to be channeled through institutions and forums of scholarly communication within which formal and informal relationships, political pedigree, and personal stamina often played roles as important as the role of the attitudes of the party’s cultural or agitprop bodies. But this also means that the escalation of an academic affair usually was fueled by a certain stubbornness or hard-minded attitude on the part of those who were targeted by the party authorities for political reasons. This kind of stubbornness certainly played a vital role in the case of Hegedüs, Kemény, and Szelényi. Yet their work was hardly intended initially as an immediate challenge to ideological or political barriers. Their dogged determination stemmed rather from their professional commitment to the value of sociological research, which, due to the more and more unsound and ambiguous standards of scholarly performance introduced for ideological reasons, gradually morphed into the adoption of a stance which could rightly be called oppositional, although in each case with a different connotation.

The fact that the revitalization of sociological research and the launch of empirical investigations were closely connected to economic reform drives from the mid-1960s had some serious consequences for the fate of sociology as a discipline in Hungary. First of all, as was made explicit by the case of Hegedüs, the invasion of Czechoslovakia by the Warsaw Pact countries in 1968 and the subsequent halt of the reforms were perceived by many as a defeat and as a consequent shrinking of scholarly opportunities. Yet what really counted was not necessarily the political face-value of these events. It was, rather, the lack of a positive model under these circumstances for valuable and pioneering sociological research which affected negatively the academic performance and vision of progressive sociologists. Whereas the Guidelines on Scientific Policy issued by the Central Committee of the party in 1969 accorded unlimited liberty to research in social science, it also called for “prudence and responsibility” (i.e. self-censorship) in making scientific results available to the public.71 But the nature of empirical findings in sociological research and the flexible and institutionally different understanding of scientific responsibility rapidly revealed the ideological frailty of these claims. Especially in sociology, in which scientific truth is supposedly founded on the critical observation of social facts, any demand for self-control and self-limitation could produce utterly counterproductive if not false results. Combined with the conviction that politics can discover in sociology something which it cannot discover by any other means, sociological responsibility in principle overtly fostered the emergence of a critical attitude.

Thus, one can understand why Hegedüs, even after his many conflict-ridden entanglements with the political world in Hungary, kept stubbornly challenging the prevailing view on socialist development in Hungary, arguing that the return to market conditions was in fact a false turn, because it intensified social stratification and inequality. The adoption of a political stance in this case was clearly motivated by sociological insight into society and its amalgamation and a belief in the idea of a genuinely socialist democratization of human relations. Similarly, Kemény’s uncompromising excavation of delicate social facts was linked to his belief in the unconditional value of the empirical study of social reality, even if it had regularly culminated in sociological analyses touching critically on some basic ideological tenets. Finally, Szelényi’s increasingly radical approach to sociology as a critique of ideology originated in and was founded on his perception of the discrepancy between certain empirically detected social tendencies which fostered inequality and a particular set of socio-politically promoted interests in society which supported them. In each of these cases, the only legitimate option offered by the academic establishment for sociological work consisted of keeping a low profile from the perspective of critical attitude and promoting the very social status quo the shortcomings of which had been revealed by sociological means. No wonder that for each of the three scholars irritation and disappointment with this situation, which was also for them a sociologically reflected disposition, called for a radical response: the emergence of an oppositional attitude both in their scholarly work and in the ways in which they were more and more ready to take serious political risks.

Bibliography

Primary Sources

 

National Archives of Hungary (MNL OL)

M-KS 288-5. Politikai Bizottság [Politburo], 1956–1989.

M-KS 288-41. Agitációs és Propaganda Bizottság, 1962–1989

 

Az MSZMP Központi Bizottságának Tudománypolitikai Irányelvei [The Guidelines for Scientific Policy of the Central Committee of the HSWP]. Budapest: Kossuth Kiadó, 1969. 37–67.

“A Magyar Szocialista Munkáspárt néhány időszerű ideológiai feladata: A Központi Bizottság irányelvei” [Some current ideological tasks of the Hungarian Socialist Workers’ Party: The guidelines of the Central Committee]. In A Magyar Szocialista Munkáspárt dokumentumai és határozatai 1963–1966 [The documents and resolutions of the Hungarian Socialist Workers Party 1963–1966], edited by Henrik Vass. Budapest: Kossuth, 1970.

“Az MSZMP Központi Bizottsága mellett működő Kultúrpolitikai Munkaközösség állásfoglalása néhány társadalomkutató anti-marxista nézeteiről” [The position of the Cultural Political Work Collective operating along with the Central Committee of the Hungarian Socialist Workers’ Party on the Anti-Marxist views of several social researchers]. In A magyar demokratikus ellenzék. vol. 3. Dokumentumok [The Hungarian democratic opposition, vol. 3. Documents], edited by Sándor Csizmadia. Budapest: T-Twins, 1995.

 

Secondary Sources

 

Apró, Éva. “Mi késleltette a magyar városfejlődést?” [What slowed down Hungarian urban development?]. Társadalmi Szemle 27, no. 6 (1972): 28–38.

Csizmadia, Ervin. A magyar demokratikus ellenzék (1968–1988) [The Hungarian democratic opposition (1968–1988)]. Budapest: T-Twins Kiadó, 1995.

Hegedüs, András. “Optimalizálás és humanizálás: Az igazgatási rendszer modernizációjáról” [Optimization and humanization: On the modernization of the system of administration]. Valóság 9, no. 3 (1965): 17–32.

Hegedüs, András. A szocialista társadalom struktúrájáról [On the structure of socialist society]. Budapest: Akadémiai Kiadó, 1966.

Hegedüs, András. A szociológiáról: Egy tudomány lehetőségei és korlátai [On Sociology: The potentials and limitations of a science]. Budapest: Akadémiai Kiadó, 1966.

Hegedüs, András. “Realitás és szükségszerűség: A szocialista társadalom ’önbírálata’, mint realitás és mint szükségszerűség” [Reality and necessity: The self-criticism of socialist society]. Kortárs 11, no. 7 (1967): 1011–19.

Hegedüs, András. “A marxista szociológia egészséges fejlődéséért!” [For the healthy development of Hungarian sociology!]. Társadalmi Szemle 23, no. 10 (1968): 93–99.

Hegedüs, András. Socialism and Bureaucracy. London: Allison & Busby, 1976.

Hegedüs, András. Élet egy eszme árnyékában [Life in the shadow of an idea]. Budapest: Bethlen Gábor Könyvkiadó, 1989.

Hegedüs, András. “Beszélgetés” [Conversation]. In Hatalom, politika, társadalomtudomány: Interjúk a magyar szociológia újjászületésének körülményeiről az 1960-as években [Power, politics and social science: Interviews on the circumstances of the rebirth of Hungarian sociology in the 1960s], edited by Tamás Rozgonyi, 6–17. Budapest: HAS Sociological Institute, 2008.

Herceg, Ferenc. “A szocialista elosztás néhány kérdése” [Some questions of socialist distribution]. Társadalmi Szemle 26, no. 1 (1971): 65–72.

Huszár, Tibor, et al., eds. Hungarian Society and Marxist Sociology in the Nineteen-Seventies. Budapest: Corvina, 1978.

Keen, Mike, and Janusz Mucha, eds. Eastern Europe in Transformation: The Impact of Sociology. Westport: Greenwood Press, 1994.

Keen, Mike, and Janusz Mucha. Autobiographies of Transformation: Lives in Central and Eastern Europe. London: Routledge, 2006.

Kemény, István. “Az úton lévők hatalmas tábora” [The huge camp of those on their way]. Népszava, October 17, 1969.

Kemény, István, ed. Beszámoló a magyarországi cigányok helyzetével foglalkozó 1971-ben végzett kutatásról [Report on the research dealing with the situation of the Roma population in 1971]. Budapest: HAS Sociological Institute, 1976.

Kemény, István. Ouvriers hongrois. Paris: L’Harmattan, 1985.

Kemény, István. Szociológiai írások [Sociological writings]. Budapest: Replika Könyvek, 1992.

Kemény, István. Velük nevelkedett a gép: Magyar munkások a hetvenes évek elején [The machines grew with us: Hungarian workers at the beginning of the 1970s]. Budapest: Vita, 1990.

Kemény, István. “Poverty in Hungary.” Social Science Information 18 (1979): 247–67.

Kemény, István, and László Gábor, eds. XXX: 1963-ban alakult meg a Szociológiai Kutatócsoport [XXX: In 1963 the Sociological Research Group was founded]. Budapest: Akaprint, 1994.

Kemény, István, and Béla Janky. “The Roma Population in Hungary 1971–2003.” In Roma of Hungary, edited by István Kemény, 70–225. Boulder: Atlantic Research and Publication Inc, 2005.

Kemény, István. “Interview on his career conducted by Vera Szabari.” Sociological Review, 18, no. 2 (2008): 135–54.

Konrád, György, and Iván Szelényi. Az új lakótelepek szociológiai problémái [Sociological problems of new urban settlements]. Budapest: Akadémiai Kiadó, 1969.

Konrád, György, and Iván Szelényi. “A késleltetett városfejlődés társadalmi konfliktusai” [The social conflicts of delayed urban development]. Valóság 15, no. 12 (1971): 19–35.

Konrád, György, and Ivan Szelényi. The Intellectuals on the Road to Class Power, translated by Andrew Arato and Richard E. Allen. New York–London: Harcourt Brace Jovanovich, 1979.

Mód, Aladárné, et al. Társadalmi rétegződés Magyarországon [Social stratification in Hungary]. Budapest: Central Statistical Office, 1966.

Nemes, Ferenc, and Iván Szelényi. Lakóhely és közösség [Settlement and Community]. Budapest: Akadémiai Kiadó, 1967.

Nyers, Rezső. “Emlékeim Hegedüs András pályafutásának három korszakáról” [Memories on the three Stages of the career of András Hegedüs]. In Búcsú Hegedüs Andrástól 1922–1999 [Farewell to András Hegedüs, 1922–1999], edited by Tamás Rozgonyi and Zoltán Zsille, 259–62. Budapest: Osiris, 2001.

Szántó, Miklós. A magyar szociológia újjászervezése a hatvanas években [The reorganization of Hungarian sociology in the 1960s]. Budapest: Akadémiai Kiadó, 1998.

Szelényi, Iván. “Empíria és szociológia” [The empiric and sociology]. Valóság 12, no. 11 (1969): 14–26.

Szelényi, Iván. Urban Inequalities under State Socialism. Oxford: Oxford University Press, 1983.

Szelényi, Iván. Városi társadalmi egyenlőtlenségek [Urban social inequalities]. Budapest: Akadémiai Kiadó, 1990.

Szelényi, Iván. “Utószó: Jegyzetek egy szellemi önéletrajzhoz” [Afterword for an intellectual self-portrait]. In idem. Új osztály, állam, politika [New class, new state, new politics]. Budapest: Európa Kiadó, 1990.

Szelényi, Iván. “Nosztalgikus jegyzetek a hatvanas évekről” [Nostalgic notes on the 1960s]. Mozgó Világ 41, no. 4 (2015): 80–97.

Szelényi, Iván. “Egy kézirat története” [The story of a manuscript]. Kritika 45, no. 7–8. (2015): 4–10.

Tőkés, Rudolf L. Hungary’s Negotiated Revolution: Economic Reform, Social Change and Political Succession. Cambridge: Cambridge University Press, 1996.

Walter Pohl

Director, Institute for Medieval Studies, The Austrian Academy of Sciences

 

The paper deals with the significance of ethnonyms for the study of early medieval ethnicity. The historiographic sources are full of names of peoples, and endow them with collective agency. That may not prove that all of these peoples had strong ethnic identities. But it attests to the general use of ethnicity as a cognitive device to differentiate between large social groupings who were relevant actors on the political scene. In this scheme, ethnonyms are fundamental. ‘Ethnicity’ as a system of distinctions between collective social actors and ‘ethnic identity’ as the result of a series of identifications are of course closely linked, but they represent different aspects of ‘the ethnic’. Therefore, ethnonyms do not necessarily reflect ethnic self-identification of the group concerned, although they often do. What they attest to is some shared belief that humans can be distinguished by ethnonyms, that is, on the basis of ‘natural’ affiliations that people are born with.       

Keywords: ethnonyms, early medieval ethnicity, Longobards, Goths, gentes

What did ethnonyms mean in the early medieval period?1 We can begin with an example of what people thought about this question themselves. In the middle of the seventh century, the origin story of the Longobards was written down in the Longobard kingdom in Italy in a text called Origo gentis Langobardorum. Toward the end of the eighth century, Paul the Deacon faithfully repeated the story in his Historia Langobardorum, although he (a Christian monk) distanced himself from it by calling it a ridicula fabula.2 According to these two texts, a long time ago a small people called the Winnili migrated from Scandinavia, led by the wise woman Gambara and her sons. They were challenged by the Vandals, who solicited the support of Wodan (a Germanic god of war). Gambara therefore asked Wodan’s wife Frea for support, and she gave the advice that the Longobard women should tie their hair in front of their faces so that it looked like a beard and go with the men to the battlefield. When Wodan awoke the next day, he looked out on the battlefield and asked “Who are these longbeards?” Frea replied, “As you have given them the name, give them victory!” From then on, the Winnili were called Longobards.3

In all likelihood, this is a pre-Christian story based on the supposed agency of pagan gods.4 If Wodan gives a name to the people, he adopts it in a sense, and is obliged to give it victory. Scholars have long assumed that Wodan had (unwittingly, as the legend implies) conferred one of his own epithets on the Winnili. Yet the fourteenth-century text in which langbardr is listed among Odin’s/Wodan’s names may also have relied on a knowledge of Paul the Deacon’s Historia.5 The name conferred on the Longobards is strikingly paradoxical. As Paul the Deacon states, “it is certain that the Longobards were afterwards so called on account of the length of their beards untouched by the knife.”6 This is a rather straightforward explanation, immediately comprehensible both in Germanic languages and in Latin (longibarbi, as Wodan says in Paul’s account). It was also taken up by Isidore of Seville in his seventh-century Etymologies: Langobardos vulgo fertur nominatos prolixa barba et numquam tonsa, “the Longobards, according to popular opinion, are named after their long beards that are never cut.”7 However, the origin story subverts this clear-cut etymology based on a secondary male sexual characteristic by attributing the long beards to women, and the narrative privileges female agency: Gambara, as leader of the Longobards, is more successful by relying on Frea, than the Vandals, who have directly appealed to Wodan.8 Whatever the implications of this and other stories about “women in the beginning,”9 this narrative must have allowed the women to regard themselves as Longobards in the full sense, too. This, then, is a story of self-identification with and through an ethnonym.

At the same time, ethnonyms also allowed external identification of peoples. This is illustrated by a second example from a somewhat earlier period. The Historia Augusta, written around 400 AD, offers a detailed and fictive description of the Emperor Aurelian’s triumph, thought to have taken place in the 270s. According to this account, Aurelian rode up to the Capitol in a chariot which had belonged to a king of the Goths and was drawn by four stags, followed by exotic animals, gladiators, and captives from the barbarian tribes, among them Arabs, Indians, Persians, Goths, Franks, and Vandals. “Ten women were also led along, who, fighting in male attire, had been captured among the Goths after many others had been killed; a placard declared these women to be of the kin [genus] of the Amazons—for placards are borne before all, displaying the names of their people [gens].10 This set-up (representatives of a people marching past the spectators, one after the other, carrying signs with their names) reminds one of the grandiose opening ceremonies of the Olympic Games of our time. At Roman triumphs and in panegyrics devoted to Roman emperors, long lists of defeated peoples were a standard feature. Again, the functional logic of ethnonyms is somehow subverted by women: the Amazons, a fictive female people. The Gothic women found fighting on the battlefield certainly did not constitute a people of their own; but their spectacular presence in Aurelian’s triumph was endorsed by ancient mythology.11

In both examples, ethnonyms are a central feature of ethnic identification. Contemporaries tended to believe that they represented the nature of a people, an assumption that Isidore of Seville systematically employed in his Etymologies to explain the characteristics of the numerous peoples that he lists.12 Indeed, some names carried a clear meaning in the language of their own people, such as Longobards or Alamanni (“all” or “full” men). Others, mostly by coincidence, could easily be (mis)understood in Latin: Saxons (rocks), Angli (angels), Bulgars (vulgar), or Avars (greedy). The names already seemed to tell a story, as in the Longobard origin myth.

Ethnonyms, furthermore, were the usual way to structure the political world, and the history of its changes. Some texts (judging from the manuscripts) bore the names of peoples in their titles, if in rather different phrasings: De origine actibusque Getarum, Origo gentis Langobardorum, Liber Historiae Francorum, Historia ecclesiastica gentis Anglorum. Still, the ethnonyms were not without ambiguities. In the construction of the Getica (which by the way is a modern title), the Goths were identified with the ancient Scythians and Dacians, and in particular, with the similarly-named Getae, who were referenced in the title. The intention was to enhance the ancient glory of the Goths, but this created rather confusing equations. Isidore, in his Gothic history, proposed the rather far-fetched argument that the names were so similar that “with one letter removed and one letter changed, ‘Getae’ becomes ‘Scythae’.”13 Isidore also added an identification with the apocalyptic peoples of Gog and Magog, featured in the prophecies of Ezekiel and in John’s Apocalypse: Gothi a Magog filio Iaphet nominati putantur, de similitudine ultimae syllabae “the Goths are supposed to be named after Magog, son of Japhet, because of the similitude of the last syllable.”14

What historians habitually refer as “Anglo-Saxons” or simply the “(early) English” was in fact a conglomerate of peoples, mainly Angles and Saxons; in different passages, Bede variously adds Jutes and/or Frisians, and sometimes other names.15 Bede did much to promote the name Angli/English for all of them, not least because of the association with angels, expressed in a famous saying attributed to Pope Gregory the Great: “Not Angles, but angels” (whereas “Saxons” could be understood as “stones” or “daggers”). And the Longobards only got their name at the beginning of their written history; in later manuscript catalogues, Paul the Deacon’s History of the Longobards was occasionally still entered as “History of the Winnili.” This does not mean that these peoples had no solid identities, and in a sense the onomastic multiplicity could also enhance their pride. It does however indicate that these stories were about identities in the making, not about clear-cut routines of identification.

The ethnic element of identity is prominent in the early medieval sources because, at the end of Antiquity, the countries mostly came to be named after the people by which they were inhabited , and not vice versa. Gaul became France, a large swathe of the ancient province of Liguria came to be Lombardy, the main part of Britain, England. Later, what had been Pannonia became Hungary; instead of Thrace, there was Bulgaria; and northwestern Illyricum became Croatia. Only the Goths did not reign long enough to leave their name on their former realms. In the long run, some of the ancient territorial designations in Europe were maintained: Italy, Spain, Britain, Greece/Hellas (Belgium was only re-appropriated by the new state in 1830), and some regional names such as Aquitaine, Tuscany, Dalmatia, and Macedonia. A few new territorial designations appeared over the course of history, for instance Castile, Provence, Lotharingia/Lorraine, the Netherlands, Switzerland, and Austria. Northern and eastern Europe, beyond the former Roman borders, have an almost exclusively ethnic topography: Sweden, Denmark, Finland, Poland, the Czech Republic, Serbia, and Russia.

This ethnic configuration of the political geography of large parts of Europe was not a straightforward development. Ethnicity was not necessarily the prime mover of medieval and early modern European history. It mattered more or less as a form of identification and social cohesion, depending on the circumstances. If the names of states and peoples on today’s map of Europe are surprisingly similar to those on a map from one thousand years ago, this is not because these peoples and states had unbroken histories of linear development. Some disappeared from the map for centuries (for instance Poland, Bulgaria, and Serbia), or their geographical position shifted (for instance Burgundy, Bulgaria, and Lithuania), or they were conquered by foreigners (for instance England by the Normans and much of southeast Europe by the Ottomans) or lived through periods of fragmentation (for instance France and Germany). For a long time, Francia was only the core of the Frankish realm, more or less today’s Île de France. But even where political independence or continuity of a sense of ethnic community were interrupted, they remained available to later appropriations. Sometimes such appropriations were fictive, and rested on the similarity of the name or of the region. For instance, the “Wends” (a German name for the Slavs) were soon identified with the long-disappeared Vandals, a self-representation which reached its peak in the late medieval and early modern period.16

What remained in place throughout all these changes was the principle of a distinction by ethnonyms. In this simple sense, “ethnicity” is a system of distinguishing between named social groupings according to their ethnonyms and ascribing collective agency to them.17 For the early Middle Ages, we have only patchy information about ethnic self-identification. However, we have ample evidence for the systematic employment of ethnic distinctions, mostly by outside observers, as shown in the example from the Historia Augusta. In that sense, the early Middle Ages were a world of gentes. In the narrative sources, collective agency was unproblematically attributed to peoples: they migrated, converted to Christianity, waged war, or raised kings. A state or a kingdom could hardly act as a collective; it was only the king as the representative of the people or the people itself who could take political action. Ethnic agency also applied to smaller groups and non-state actors, as long as they could be identified (otherwise such groups would often be generally labeled “barbarians”). In that sense, ethnicity was generally used as a system of distinctions between gentes which made it possible to structure the social world and to circumscribe collective political actors and broad, inclusive social groups. This raises problems of definition: can we distinguish between ethnic and other social groups, or do they represent a continuum in early medieval usage? And what distinguishes an ethnonym from a territorial or political label?

It is hard (and controversial) to define “ethnic.”’ in an unambiguous way. Many scholars offer definitions with lists of distinctive features (common origin, memories, language, culture, customs, costume, territory, etc.).18 These kinds of definitions mostly apply to urban or territorial identities as much as they do to ethnic identities. There are also subjective definitions, according to which ethnic identity is determined by a subjective sense of belonging to a group.19 However, we have relatively little evidence of actual subjective self-definitions in the early Middle Ages. Therefore, I would propose four answers to the question of definition.

First, we all know in everyday usage what an “ethnic” name is, and so did ancient and medieval historians. Our understanding obviously differs little, for instance, from the one laid out by Isidore of Seville in his Etymologies in the seventh century, which remained popular throughout the Middle Ages. Most of the nomina gentium which he lists (with etymological explanations) are also ethnonyms by our standards.20 They include the Romans (at the time often considered one gens among others), but otherwise only a few groups that we might not consider as ethnic. Isidore also discusses the terminology (gens, natio) and the relationship between peoples and languages. The unquestioned assumption is that after the Flood, the world was divided up by gentes according to their descendance from the sons of Noah.21 Consequently, Isidore defines gens as a multitude descended from one origin, but he then adds an alternative: “or distinguished by its particular grouping.”22 The twenty books of his Etymologies contain only two other chapters which provide exhaustive lists of named social groups: imaginary peoples (the “monstrous races,” that is, fantastic ethnicity) and Christian heresies (often named after their founder, for instance “Arianism”).23 Isidore was surely able to draw the line between gentes and other social groups. The same applies to the general historiographical use of ethnonyms, most of which seem to correspond to modern notions of ethnicity.

Second, as noted above, according to ancient and medieval perceptions, countries and polities cannot act, only people and their representatives can. In our political language, Washington or France can take political action. Rome or the regnum of the Franks do not have agency, only the senatus populusque Romanus or the rex and the gens Francorum can act. The ancient populus essentially implied a political definition of the “people,” not an ethnic one. In the ancient period, the notion of civic identity was so strong that the populus, the people of a city, dominated the political landscape and the historical narratives. In the early Middle Ages, this changed, and the gentes came to the fore. Thus, the Romans came to be regarded as one gens among many.24 Still, there are some cases in which the texts also attribute the same kind of agency to groups that we might not regard as ethnic, for instance the “Romans” of the eastern Roman Empire (who by our standards were mostly Greeks) or the populations of cities (for instance the Venetians), (former) provinces (the Aquitanians), and smaller kingdoms (the Mercians). Our more neutral term “peoples” may thus be more appropriate to cover the entire range of collective agents.

This leads to the third element of definition: on a pragmatic level, an ethnonym is defined by its position in a horizontal system of distinctions within the social world. If the prevalent distinction is between gentes, then named collective actors whom we would not regard as ethnic groups (Romans, Normans, or Venetians, for instance) tend to be ethnicized as well, and can be presented in the texts as a gens Romanorum, Normannorum, and Veneticorum.

The fourth element of a terminological clarification tends to be narrower. The term gens, which is overwhelmingly used for early medieval peoples, comes from gignere, to procreate; genus and natio have a similar etymological background. This suggests that gentes were understood as “having a common origin,” regardless of whether or not the people in question actually did. In this context, “ethnic,” in my view, can most usefully be understood as a perceived intrinsic quality that is in the people themselves: common blood, common origin, or a similar quality. Thus, it needs no defining point of reference outside the person, such as a city, a land, a polity, or a religious cult.25 One can be a Goth or a Hun wherever one is, under Hunnic, Gothic, or Roman rule, as a pagan or a Christian. Of course, in most cases ethnic identities attach themselves to territorial, political, religious, or other identities and form amalgamates of identification. Yet it is methodologically more advantageous to be able to distinguish between these different elements of identification in order to analyze how their relative significance changes. For instance, is the affiliation with the people crucial, or is the affiliation with the land more important? It makes a difference whether a royal title is rex Hungarorum or rex Hungariae. However, it is not a fundamental difference (the land is named after the people), but a gradual one.

The approach defined by these four methodological principles is necessarily flexible. It cannot rely on one clear definition which can be used for all periods, but compels us to historicize our concepts. The goal is not to decide whether or not an early medieval people “was” an ethnic group. That would be a static and not very productive approach. Three questions may be more interesting. One is the question of the extent to which a people or peoples in general were regarded by contemporaries in ways that fit our criteria for ethnicity. The second is the question of our heuristic purposes to use the concepts of ethnicity and ethnic identity. Thirdly, this gradual approach allows us to assess how the salience and meaning of ethnicity changed over time or differed in different contexts at the same time.

This flexible approach also allows us to deal with a good number of problematic cases of ethnonyms. First, some ethnonyms found in biblical, ancient, or medieval sources are clearly fictive. But as argued above, educated observers could basically distinguish between actual people and “monstrous races.” As we have seen, Isidore draws clear distinctions between them.26 Second, frequently ancient and outdated names were used, which were sometimes conjured up to make the victories of a Roman emperor seem more impressive or, in other cases, to refer to ethnographic stereotypes or relatively stable identifications of earlier with later peoples. Thus, the Huns could be called Scythians, the Avars Scythians and Huns, and the Hungarians by all of these names.27

Third, names employed by outsiders could consistently differ from the name used for self-identification. This can be an enormously stable practice: the Hellenes have been called “Greeks” by many of their neighbors for more than 2000 years, and they still are. If such a case of cultural translation is well-established and generally known, it may create surprisingly few problems. The modern Deutschen are called Germans by the English, Allemands by the French, Tedeschi by the Italians, Němci (or something similar) by many Slavic peoples, and Saksa by the Finns, but everybody seems to be well aware who is who.

Names for collectives only mentioned in isolated texts may not help much to establish any “real” identities. More frequent mentions at least allow one to trace consistent naming practices within a wider system of distinction. It may still be difficult to grasp to what extent this mental map corresponded to social practice, or in this case, to an ethnic identity. A decisive criterion is whether there is evidence to suggest interaction and communication between the author of the source, his environment, and the people in question. In general, the representatives of the Roman, Byzantine, or Carolingian empires could hardly afford to deal with their many neighbors on the basis of totally fictitious mental maps. Some inconsistencies are always noticeable, especially in the barbarian lands and the steppe zone; in many cases, they may point to shifting identifications. At almost the same time, around 550, both Jordanes and Procopius provided a generally consonant, but to some extent contradictory map of peoples living around the Black Sea.28 East Roman diplomats and travelers provided the material for these kinds of ethnographic descriptions. The contact with Romans may even have convinced some smaller peoples in the area that they were in fact Scythians or Huns.

Byzantine name-giving, according to Florin Curta’s hypothesis, gave the impulse for the spread of the name “Slavs.”29 As I have argued, at least in the Latin West, the name “Slavs” came from Constantinople, not from communication with the Slavs themselves.30 We can trace the way in which the use of the name spread, for instance through a letter of the exarch of Ravenna, who informed Pope Gregory I, who had previously only spoken of “barbarians.” John of Biclaro, who had spent many years in Constantinople, introduced it in distant Spain. Frankish authors only employed it in the seventh century. The European Avars, ridiculed as “pseudo-Avars” by the Byzantines, were supposed to have soon adopted a prestigious name given them by other peoples; Turks and Byzantines initially called them “Varchonites.”31 The Byzantines very insistently called the Magyars/Hungarians “Turks,” and even sent a golden crown to the Hungarian king with the inscription “kralēs Tourkias,” King of Turkey; but this never turned into a self-designation.32

These and similar examples should not be used in support of the claim that ethnic identities were infinitely malleable and did not really matter. Ethnicity mattered, not least because it was controversial and not easy to handle on a conceptual level. It was always a matter of communication and cultural translation, and a way of placing oneself and one’s own community within a wider world of gentes. This ethnic landscape was constantly changing, but at the same time, it also provided a familiar long-term perspective for identifications. Most ethnonyms that one finds in early medieval sources were used for considerably longer than an individual lifetime. They made the world more predictable, in part because the names and some of the background information connected with the respective peoples hinted at what one could expect from them.

Given the evidence that we have, then, ethnicity can most easily be studied on the discursive level as a way of structuring the social world and of ascribing agency to broad social groups. In pre-modern societies, there were not many levels on which the naming of macro-groups was so systematically pursued. In many historical contexts, ethnonyms and a very culture-specific terminology of peoplehood shaped perceptions of large groupings and guided political decisions. For instance, it made a big difference in Late Antiquity whether groups immigrating from beyond the Roman border were perceived simply as unspecified “barbarians” or were identified using ethnic distinctions (which made it possible to play them off against one another and to rely on previous experiences with the same or similar groups). Apart from serving as a cognitive tool, ethnic discourse also provided a powerful framework with which to express “visions of community.” It could be extended far beyond the range of groupings that we would describe as “ethnic,” at least in metaphorical ways. For instance, as Denise Buell has shown, the early Christians could be described in ethnic terms.33 That makes the concept of ethnicity hard to delineate and define. We would hesitate to class Christians as an ethnic group. On the other hand, such uses indicate the potential of ethnic language to promote social cohesion or, indeed, disruption. It is the very success of ethnicity in many historical contexts that makes the concept fuzzy for scholarly uses. Yet this is the challenge that makes research on ethnicity so interesting.

 

Bibliography

Primary Sources

Bede. “Historia Ecclesiastica gentis Anglorum.” In idem: Opera historica, edited and translated by J. E. King, 2 vols. Loeb Classical Library. London: Heinemann, 1930.

Chastagnol, André, ed. Historia Augusta. Paris: Robert Laffont, 1994.

Iordanes. Getica. Edited by Theodor Mommsen. Monumenta Germaniae Historica, Auctores Antiquissimi 5.1. Berlin: Weidmann, 1882.

Isidore of Seville. Etymologiae. Edited by Wallace Martin Lindsay. Oxford: Oxford University Press, 1911.

Isidore of Seville. “History of the Goths.” Translated by Kenneth Baxter Wolf. In Conquerors and Chroniclers of Early Medieval Spain [TTH9], 79–109. Liverpool: Liverpool University Press, 1999.

Origo gentis Langobardorum. Edited by Georg Waitz. Monumenta Germaniae Historica, Scriptores rerum Langobardicarum et Italicarum saec. Vols 6–9. Hannover: Hahn, 1878.

Paul the Deacon. Historia Langobardorum. Edited by Georg Waitz. Monumenta Germaniae Historica, Scriptores rerum Langobardicarum et Italicarum saec. Vols. 6–9. Hannover: Hahn, 1878. [English translation: History of the Lombards. Translated by William Dudley Foulke. Philadelphia: University of Pennsylvania Press, 1907.]

Procopius. Bella. Edited by H. B. Dewing. Loeb Classical Series. 5 vols. Cambridge, Mass.: Harvard University Press, 1914–1928.

 

Secondary Sources

Buell, Denise K. Why this New Race? Ethnic Reasoning in Early Christianity. New York, NY.: Columbia University Press, 2005.

Curta, Florin. The Making of the Slavs: History and Archaeology of the Lower Danube Region c. 500–700. Cambridge Studies in Medieval Life and Thought: Fourth Series. Cambridge–New York: Cambridge University Press, 2001.

Geary, Patrick J. Women at the Beginning: Origin Myths from the Amazons to the Virgin Mary. Princeton: Princeton University Press, 2006.

Liccardo, Salvatore. “Different Gentes, Same Amazons: The Myth of the Women Warriors at the Service of Ethnic Discourse.” In Narratives of Ethnic and Tribal Origins, edited by Walter Pohl, and Daniel Mahoney. Special issue, Medieval History Journal (forthcoming, 2018 Autumn).

Nedoma, Robert. “Der altisländische Odinsname Langbardr: ‚Langbart‘ und die Langobarden. In Die Langobarden: Herrschaft und Identität. (Forschungen zur Geschichte des Mittelalters), edited by Walter Pohl, and Peter Erhart, 439–44. Vienna: Verlag der Österreichischen Akademie der Wissenschaften, 2010.

Pohl, Walter. “Narratives of Origin and Migration in Early Medieval Europe: Problems of Interpretation.” In Narratives of Ethnic and Tribal Origins, edited by Walter Pohl and Daniel Mahoney. Special issue, Medieval History Journal (forthcoming Autumn 2018).

Pohl, Walter. “Gender and Ethnicity in the Early Middle Ages.” In Gender in the Early Medieval World: East and West, 300–900, edited by Leslie Brubaker, and Julia Smith, 23–43. Cambridge: Cambridge University Press, 2004.

Pohl, Walter. “Aux origines d’une Europe ethnique: Identités en transformation entre antiquité et moyen âge.” Annales: Histoire, Sciences sociales 60, no. 1 (2005): 183–208.

Pohl, Walter. “Ethnic Names and Identities in the British Isles: A Comparative Perspective.” In The Anglosaxons from the Migration Period to the Eighth Century: An Ethnographical Perspective, edited by John Hines, 7–40. Woodbridge: Boydell Press, 1997.

Pohl, Walter, and Philipp Dörler. “Isidore and the gens Gothorum.” In Isidore de Séville et son temps. Antiquité Tardive 23 (2015): 133–42.

Pohl, Walter. Die Awaren: Ein Steppenvolk in Mitteleuropa, 567–822 n. Chr. Munich: C.H. Beck, 2015 [reworked English translation forthcoming: The Avars: A Steppe Empire in Central Europe, 567–822. Ithaca: Cornell University Press, forthcoming 2018].

Pohl, Walter. “Introduction: Strategies of identification. A methodological profile.” In Strategies of Identification: Ethnicity and Religion in Early Medieval Europe, edited by Walter Pohl and Gerda Heydemann, 1–64. Turnhout: Brepols, 2013.

Pohl, Walter. “Huns, Avars, Hungarians – Comparative Perspectives Based on Written Evidence.” In The Complexity of Interaction along the Eurasian Steppe Zone in the First Millennium CE. Contributions to Asian Archaeology, vol. 7, edited by Jürgen Bemmann and Michael Schmauder, 693–702. Bonn: Rheinische Friedrich-Wilhelms-Universität, 2015.

Steinacher, Roland. “Wenden, Slawen, Vandalen: Eine frühmittelalterliche pseudologische Gleichsetzung und ihre Nachwirkungen.“ In Die Suche nach den Ursprüngen: Von der Bedeutung des frühen Mittelalters. Forschungen zur Geschichte des Mittelalters 8, edited by Walter Pohl. Denkschriften der philosophisch-historischen Klasse 322, 329–353. Vienna: Österreichische Akademie der Wissenschaften, 2004.

Smith, Anthony D. The Ethnic Origins of Nations. London: Wiley-Blackwell, 1986.

Transformations of Romanness in the Early Middle Ages: Regions and Identities. Edited by Clemens Gantner, Cinzia Grifoni, Walter Pohl, and Marianne Pollheimer. Berlin–New York: de Gruyter, forthcoming 2018.

Wenskus, Reinhard. Stammesbildung und Verfassung: Das Werden der frühmittelalterlichen gentes. Cologne–Vienna: Böhlau, 1977).

Wolfram, Herwig. “Origo et Religio: Ethnic Traditions and Literature in Early Medieval Texts.” Early Medieval Europe 3 (1994): 19–38.

Odile Kommer, Salvatore Liccardo, Andrea Nowak

Odile Kommer: Institute for Social Anthropology at the Austrian Academy of Sciences
Salvatore Liccardo: Institute for Medieval Research at the Austrian Academy of Sciences
Andrea Nowak: Department of Middle Eastern Studies at the University of Vienna

This article1 examines ethnonyms for Persians in Medieval Latin, Greek, and Arabic sources. These ethnonyms are part of ethnic terminologies which changed over time and varied in different regional contexts. The ethnonyms for Persians are approached in different textual genres from a combination of historiographical, philological, and social anthropological perspectives. In the first part, the investigation of Persians in Late Antique source material sets out from the Tabula Peutingeriana and examines the entries on the map which refer to the Persians, highlighting both their ethnic and political meanings. The second part deals with source material on medieval South Arabia. First, it focuses on the texts of the tenth-century Yemeni scholar al-Hamdānī and his use of a set of ethnonyms for the Persian minority population, of which each term evokes a different association. This is followed by an analysis of the early thirteenth-century account of Persian traveler Ibn al-Mujāwir, in which the roles and meanings of ethnonyms for Persians in different narrative units are discussed. This case study shows that there are interdependencies between ethnonyms and other means of identification, such as language, lifestyle, place of dwelling, kinship, descent, and the division of the world into different spatial and ideological realms. The case of the Persians illustrates how the authors under discussion used ethnonyms as part of narrative strategies which support processes of selfing and othering.

Keywords: ethnonyms, ethnicity, historical geography, Alexander narrative, Late Antiquity, (Early) Middle Ages, South Arabia, Tabula Peutingeriana, Persian

This article focuses on the study of ethnonyms in medieval sources from Mediterranean Europe and Southern Arabia, or historical Yemen, through a comparative and interdisciplinary approach. In our understanding, ethnonyms are group designations which express ethnic differentiation. Thus, the terminological distinctions of collective groups never refer to bounded ethnic categories, nor are they fixed in their application. In this article, the case of “the Persians” serves as an example of the construction of identities through the use of ethnonyms by authors with different regional, temporal, and stylistic backgrounds in their historiographical, geographical, or cartographical accounts, as well as in literary narratives from medieval Mediterranean Europe and Southern Arabia.2 Broader categories of comparison are necessary, which are representative of various academic disciplines, including history, philology, and social anthropology.3 By thoroughly examining the sources, we have identified the following often interrelated key concepts and used them as additional categories of comparison: myths, notions of space, use of terminology, and (pseudo-)etymology.

We argue that the medieval authors under scrutiny employed ethnonyms as conceptual tools, and that ethnonyms were thus made meaningful. The Arabic sources for this case study on ethnonyms for “Persians” include two historical works by the tenth-century Yemeni scholar al-Hamdānī and a travelogue by the early thirteenth-century Persian author Ibn al-Mujāwir. The Latin and Greek source material includes the Tabula Peutingeriana and literary sources from Late Antique and Early Medieval authors.

Myths, Notions of Space, and Environmental Determinism

Myths often feature elements of great narratives which meet a universal human need for the expression of particular conditions. In this sense, they can function as a code of understandings of the world. In mythical narratives, the self and the other interact, as do human and divine elements. Furthermore, mythical narratives contain a processual element, which Angelika Neuwirth calls “myth[s] in a broken form.”4 In these narrative processes, the authors employ popular literary topoi with which they provide meaningful contributions to broader discourses.5 In the context of the analysis of ethnonyms and collective processes of identification it becomes evident that the medieval authors’ narrative strategies not only include mythical features, but that these mythical features are often linked to notions of space. In their accounts, real and imagined places, the distinction between center and periphery, environmental determinism, and spaces and places of collective memory function as unifying or separating elements. For example, a people’s ethnogenesis is constructed through processes of selfing and othering, often in reference to a certain place. As we argue in this article, ethnonyms obtain their various meanings precisely in this interplay of factors.

The Biblical-Quranic founders and ancestors of the South Arabians, together with environmental and climatic conditions, are the central elements of a mythically narrated moment in which the formation process of not only a town, but a South Arabian existence is explained. Environmental determinism is the notion that the physical environment exerts a determining influence on human societies and cultures. In South Arabian mythical narratives, the influence of planetary and stellar constellations on people and climates is particularly emphasized. The notion of environmental determinism was borrowed from Hellenistic Greek discourses and has later been applied in many regions of the world. It has often been used to suggest that some peoples are more advanced than others. In the beginning of his Ṣifat jazīrat al-ˁarab, al-Hamdānī introduces the division of the world into seven “climates” (ˀaqālīm, sg. iqlīm) in accordance with the Ptolemaic idea.6 He locates Sanaa and South Arabia in the first climate and marshals different arguments to prove that the first climate is the best and, therefore, its inhabitants are also more advanced. According to the myth, the descent from Sām (Shem) through Qaḥṭān (Joktan) and Sanaa as the initial place of settlement in South Arabia are substantiated. The narrative strengthens the authenticity of the South Arabians, as well as the qaḥṭānīyūn and their South Arabian identity, by which they differentiated themselves from the North Arabians, the ˁadnānīyūn.

Although largely following Ptolemy’s view, al-Hamdānī disagrees with him concerning a climatic region named by Ptolemy after the Ethiopians (al-ḥabasha), to which Yemen (South Arabia) is also assigned.7 It is particularly the terminological designation of this area as that of the Ethiopians which al-Hamdānī rejects. South Arabia and northeast Africa competed for power for centuries. Al-Hamdānī’s use of the term al-ḥabasha refers to the territory and the dynasty of Aksum, which was a threat to the South Arabian kingdom of the Ḥimyar, and, in the third century, gained control over Yemen. In the sixth century, the ḥabasha were finally expelled from South Arabia with the help of the Sasanian army. Al-Hamdānī takes a stance not only against the subsuming of Yemenis and Ethiopians in a geographical and terminological sense, but also against the idea of shared physical and personal characteristics. Ptolemy describes the area, ranging from the equator up to the middle of the Hijaz (the western part of the Arabian Peninsula), as being extremely exposed to the sun, which causes black skin-color, dark, frizzy, and thick hair, and the (allegedly) hot or even “uncivilized” temperament of its inhabitants. Al-Hamdānī argues against this, saying that the ḥabasha are only a minority in this area and that the skin color of the inhabitants of the region varies greatly. From the perspective of skin color, some of the inhabitants of the region are in strong contrast to the ḥabasha. He identifies the inhabitants of the Arabian Peninsula and of China (al-ṣīn) as such people, located at the edge of this zone. Obviously, al-Hamdānī dislikes the idea of subsuming the South Arabians and the Ethiopians under the same climatic zone, which would imply that they were similarly affected by environmental conditions and therefore share some characteristics. African ancestry was generally associated with inferior status by Arab authors, often related to racial stereotypes.8 Both the regional history of South Arabia and the desire to see the two ethnic categories as separate motivate al-Hamdānī to make these contentions.

The necessity of drawing a distinction between the ḥabasha and the ˁarab (“Arabs”) is also expressed in a mythical tale recounted by Ibn al-Mujāwir in his Taˀrīkh al-Mustabṣir.9 According to the story, the territory of the ḥabasha was originally connected to the territory of the ˁarab through a stretch of dry land, an empty valley which reached from Suez to Bab al-Mandab. Dhū l-Qarnayn,10 the mythical hero figure Alexander the Great, then opened up Bab al-Mandab so the seawater would pour forth, flood the valley, and form the Red Sea. By creating the Red Sea, Alexander the Great intended to separate the two regions and grant each people their own territory under their own rule, so that the violent conflict between the ḥabasha and the ˁarab would finally come to an end. Alexander’s intentions notwithstanding, the ḥabasha did not cease invading South Arabia and besieging its inhabitants until much later in history.

 

 

Anthropogeography, Terminology, and the Affective Dimension of (Pseudo-)Etymology

In Late Antique and Early Medieval Latin and Greek literature, ethnonyms played a central role in authors’ sense of place. While cartographic representations of the world seem to have been uncommon,11 geographical knowledge was spread through numerous geographical treatises. Being purely textual, these works reflect an image of the world as the product of the totality of names of places and peoples. Thus, ethnonyms constitute a fundamental part of the conceptualization of space. Their importance for the Roman and post-Roman sense of place, as well as their longevity, made ethnonyms a central instrument in authors’ attempts to understand and organize a shifting ethnic landscape. Ethnonyms served both to contextualize the gentes dwelling on the periphery of the Roman world and to support coeval political agendas.

Medieval Arabic geography, more precisely the classical school of the tenth century, was primarily concerned with cartographical material which depicted the (Islamic) world. These maps were accompanied by rather short explanatory commentaries.12 Ibn Ḥawqal revised, rewrote, and expanded the literary commentary of the work of his predecessor al-Iṣṭakhrī and thereby crafted a geographical treatise of considerable breadth, the Kitāb Ṣūrat al-arḍ (The Book of the Image of the World).13 Ibn al-Mujāwir copied a section from this work for his South Arabian travelogue, in which Ibn Ḥawqal defines “the homeland of the Arabs,” diyār al-ˁarab.14 This section offers an example of how an ethnonym was used in an internal differentiation within the Islamic world. To define “the homeland of the Arabs,” the author takes into account elements of physical geography, i.e. mountains, landscapes, seas, deserts, and steppes, but also administrative districts and tribal territories. The ethnonym for “Arabs” is combined with the word diyār – diyār al-ˁarab. In the text and on the map, the word diyār is used again, but together with several tribal names to signify tribal territories. The meaning of the word diyār indicates that the sense of place is shaped by social and political interaction. Dār, the singular of diyār, means “dwelling, abode, house” and is often part of compound words that take on abstract meanings, e.g. dār al-aman “house of safety” and dār al-ḥarb “house of war.” These expressions refer to places/territories which are defined by military conflicts and peace treaties.15 In the expression diyār al-ˁarab, the ethnonym is used to represent the home of a large ethnic group that in itself is not homogenous; this home is further structured through a geographical and tribal terminology.

Moreover, ethnonyms can be loaded with stereotypical qualities and values. In such cases, ethnonyms take on an affective dimension which has an impact on the way they are used or influences social encounters with the respective group. If ethnonyms are applied in this manner, categorical projections of positive or negative ascriptions are made to the respective group. The affective substance of an ethnonym is particularly interesting from the perspective of the distinction between self-ascription and the ascription by others. Even though ascriptions by others can have neutral or positive connotations, they frequently entail negative characteristics and fuel processes of othering.16 While some ethnonyms do not have an obvious disparaging tone, many of the names used by Roman rhetoricians and historians have a strong affective value. When they refer to people’s looks, for instance in the case of the Lombards (who were given this name because of their long beards17) or way of life, for instance the Arabes Scenitae (who were given this name because they dwelled in tents18), ethnonyms can highlight the “barbaritas” of distant peoples and reinforce Roman attitudes towards non-Romans.

Fantastic pseudo-etymologies, a specialty of the author and traveler Ibn al-Mujāwir, can add affective value to ethnonyms and give new connotations to a group’s name. The author explains that the Arabs call the inhabitants of the highland of Ẓafār and those of the islands Soqotra and al-Masīra al-saḥara, “the sorcerers,” since, as he claims, al-siḥr, “sorcery,” is their innate characteristic. This attribution conveys a strong sense of otherness which stands in close relation to the theme of insularity. Perceived as self-contained worlds due to their remoteness, islands inspired all kinds of ideas about the other.19 These stories could be used to evoke a sense of normalcy and self-affirmation among the readership.20 It is also possible that in this case the author misinterpreted the Arabic designation for the South Arabian Sḥeri-speakers,21 since in Arabic the root consonants s-ḥ-r bear the meaning “magic” or “sorcery.”

Resorting to synonyms, i.e. literary or archaic versions of the same name, Latin and Greek authors could adapt their ethnic terminology according to the political and cultural climate. The use of antiquated ethnonyms to describe Late Antique gentes, as in the case of the Goths (which were often designated as Scythians or Getae), was not only a matter of style. By repeating ancient ethnic denominations, writers could flaunt their literary knowledge, but they also drew the attention of their readership to older narratives concerning the peoples in question. This literary strategy could be considered a sort of “defense mechanism.”22 In other words, it reinforced the belief that the new ethnonyms (such as “Goths”) did not prove the existence of new peoples. This rhetorical device clearly shines through Synesius of Cyrene’s speech addressed to the emperor Arcadius. In an attempt to urge the emperor to pursue a more aggressive policy against the Goths, Synesius considers the new ethnonym a forgery made by the barbarians to frighten the Romans, to make them believe that another foreign nation had sprung from the soil.23

In what follows, we show how ethnonyms, considered as conceptual tools, were used together with the above exemplified key concepts to form distinctive discourses in the particular case of the Persians. In Part I of the case study, Salvatore Liccardo analyzes the way in which Persians are portrayed on the Tabula Peutingeriana. Since the Tabula Peutingeriana represents a compendium of Greco-Roman geographical and cartographical knowledge, a study of the visual and written representations of the Persians on the map will serve to highlight both the adaptability and diffusion of ethnonyms, which shaped and supported a specific ethnic discourse or political agenda. In Part II, Odile Kommer studies how the Yemeni author al-Hamdānī applies different ethnonyms for Persians and how this relates to strategies of selfing and othering in the context of interethnic relations between the tribal majority population of the Yemen and local Persian minorities. Her contribution is based on an analysis of al-Ḥasan al-Hamdānī’s Kitāb Ṣifat jazīrat al-ˁarab and Kitāb al-Jawharatayn al-ˁaṭīqatayn al-māˀiˁatayn min al-ṣafrāˀ wa-l-bayḍāˀ, Arabic sources written in Yemen in the tenth century. Andrea Nowak examines how Ibn al-Mujāwir, in his thirteenth-century travelogue Taˀrīkh al-Mustabṣir, traces the presence of the Persians in Yemen throughout its history and along the travel route. Since the travel genre presents a rich blend of styles and topics, it provides different narrative units in which ethnonyms are charged with meaning. Furthermore, Part II offers an example of how an Arabic exonym which predominantly conveyed negative ideas about a (Persian) other later became a neutral and, eventually, positive connoted Persian self-ascription.

The Case of the Persians Part I – Late Roman Empire

According to C. R. Whittaker, the Tabula Peutingeriana is “the only certain map, in any sense that we would recognize it, to survive from antiquity (…) although preserved in a medieval copy.”24 The map was intended to represent the entire inhabited world (in Greek oἰκουμένη), from the Atlantic Ocean in the West to India in the East. Despite its impressive size (6.75 m long and 32-34 cm high), the copy in our possession is, however, incomplete, since it is missing the western extremity, grosso modo, corresponding to the west coast of North Africa, the Iberian peninsula and most of the British Isles. While the history of this copy is rather clear,25 the dating of its archetype remains a topic of heated debate. For the purpose of this article, suffice it to say that there is a certain degree of academic consensus on the dating of the last redaction of the Tabula to the Late Antique period, more specifically to the first half of the fifth century.26 Among the several thousand writings on the Tabula Peutingeriana, a handful concern the Persians. These elements of the map represent the focus of the present analysis, which aims to highlight the essential connection between sense of place, ethnographic reasoning, and imperial political discourse.

References to Persians consist of regional names, city names, and ethnonyms. Although these entries are not particularly abundant, the Persian world appears to occupy a significant place in the imagination of the mapmakers. The most visible entry, PERSIDA (10B5–11C3),27 designates a vast territory stretching from the Tigris to the Indus River. Though it is located in a somewhat peripheral area which the mapmakers knew only partially, Persia differs in no way from any other region. As in the case of the Roman provinces, the map’s coverage focuses primarily on the street network and the urban centers. The only significant difference is represented by the use of the Persian unit of itinerant distance, the parasang, instead of the Roman mile.28

Regarding the presence of other territorial names, the Tabula includes also the rubric PARRIA (11C1–11C2), indicating the region of Parthia. The size and position of this caption seem to reflect its relation to the term PERSIDA. One name, Parthia, clearly represents a subcategory of a bigger entity, Persia, which encompasses a much larger number of cities and streets. Turning one’s attention to city names, one can find the illustrious urban centers of Ctesiphon (capital of the Sasanian Empire), Ecbatana (capital of Media and subsequently one of the seats of the Parthian kings), and Persepolis (royal residence of the Achaemenes).29 Both the entry for Ecbatana and the entry for Persepolis contain a specific reference to the ethnic component of these cities. Ecbatana is called “Ecbatana of the Parthians,” and Persepolis is defined as the “Persian commercial hub.” The coexistence of Persians and Parthians on the map mirrors the ethnic reasoning of Late Antique writers, who often used Parthi and Persae as synonyms.30 Although several sources mention the shift of power from the Arsacid to the Sasanian Empire,31 in Roman accounts Persians and Parthians appear as part of the same ethnic entity, sharing customs and ethnographical stereotypes.

The ethnonym “Persian” recurs on the Tabula on two other occasions.32 Halfway between the regions of Mesopotamia and Persia, squeezed in a complex and confused fluvial system, there are the entries TROGODITI PERSI (10B5) and FLVMEIPERSI (10B5). They are examples of ethnic “double names.” Within this category fall ethnonyms composed of a known ethnic denomination and a second textual element which serves to specify the group in question. In most cases, these double names represent a particular ethnic subgroup belonging to a larger gens. For example, the entries ESSEDONES SCYTHAE (11A3) and ROXULANI SARMATE (7A5) refer to specific groupings ascribed to the broader ethnicities of Scythians and Sarmatians. In other cases, a textual element matched with an ethnic umbrella term can hint at something more than a simple subgroup. It can evoke the geography of a people’s dwelling, their way of life, their physical appearance, and their political structure, or it may even recall a literary figure.33 In the case of the entries TROGODITI PERSI and FLVMEIPERSI, the double names, placed a few centimeters away from each other, represent two groups which share the same ethnic origin: they are both considered Persians. Although one can only speculate about their exact meaning, an analysis of these entries will serve to highlight both the ethnographic knowledge and the political agenda of the authors of the map.

The inscription FLVMEIPERSI represents the most enigmatic case. As the inscription exists today, on the only surviving copy of the map, the legend is obscure. It could be that the term reflects the mapmakers’ decision to coin a neologism in order to emphasize the exotic nature of this people. Another possibility is that the ethnonym is unintelligible, because one or more different hands involved in the transmission did not understand and, therefore, did not reproduce a previously existing abbreviation. For a better understanding, it is necessary to propose a significant emendation of the inscription FLVMEIPERSI.34

Although any interpretation of this legend is simply a more or less informed conjecture, one could suppose that the term Flumei refers to an unspecified Flumen. Emended as Fluminei Persae, the inscription would mean “the Persians of the River.” This explanation has some advantages. It is close to the text and seems to reflect the location of the inscription, which is stretched out in close proximity to a watercourse. Additional perspectives can be gained by looking at other double names on the Tabula Peutingeriana that seem to allude to the specific geographical area inhabited by a given ethnic subgroup. For instance, the legend PARALOCAESCYTHAE (10A4) has been interpreted as referring to Scythians living on the coast of the Caspian Sea,35 while the inscription RVMI SCYTHAE (11A1) arguably refers to another group of Scythians dwelling near the River Rhymmus.36 Finally, the map also has the legends VAPII (1A2) and VARII (1A3), which plausibly relate to two ancient Germanic ethnonyms with their typical ending (“varii”).37 If Amsivarii and Chasuarii were the correct reading of the terms on the Tabula, these two terms would be another two ethnonyms on the map that may have been derived from the name of a river, since there is a connection between Amsivarii and the river Ems, as well as between the Chasuarii and the river Hase.38

However, the Fluminei Persae would differ slightly from the aforementioned cases, because the name is an allusion not to a specific river but to an unnamed one.39 A look at the Cosmographia of Julius Honorius,40 a geographical treatise which is roughly coeval to the Tabula Peutingeriana, might help find a more equivalent example. In one of the different catalogues which constitute this work, one finds the ethnonym Fluminenses gens.41 Based on its position in the text (after the Feratenses and the Barzufulitani, but before the Quinquegentiani) and its content, Philippe Leveau has proposed interpreting this name as referring to a specific group of Mazices, a people of Mauretania Caesariensis, which lived next to the River Chelif.42 Since the Tabula Peutingeriana and the Cosmographia are similar, both in terms of chronology and the ethnonyms employed,43 the Fluminenses could represent an analogous case to the FLVMEIPERSI and therefore support the interpretation of the map’s legend as referring to the “Persians of the river.”

On the basis of this reasoning, one could hypothesize that TROGODITI PERSI and FLVMEIPERSI were used by the mapmakers to designate two ethnic groups living in two different environments. The “Persians of the River” could represent the inhabitants of the Tigris and Euphrates river valley, while the “Persian Troglodytes” could be the dwellers of the Zagros Mountains. However, the ethnic “double name” TROGODITI PERSI carries a meaning broader than a simple geographical characterization.

The last consideration introduces a subject central to this section of this article: the analysis and contextualization of the legend TROGODITI PERSI. Albeit less obscure, this inscription is also unclear. First, the text needs a small emendation: the inclusion of an ‘l’ in the term Trogoditi, which should read Trog<l>oditi. The unusual location of the legend is also problematic. In the segments representing the eastern and far eastern lands, the depiction of both physical and urban landscapes is often inaccurate. Nevertheless, the discrepancy between the content of the inscription TROGODITI PERSI and its position is particularly striking, because of the ethnographic tradition and evocative power connected to the term “troglodytes.” While Greek and Roman geographers used this term in connection with various ethnic groups living on the fringe of the inhabited world, most frequently in Ethiopia, the authors of the map put the cave dwellers next to a meander of the Tigris, making the homeland of this people anything but peripheral. In contrast with the comparative absence of cities and roads typical of the northern periphery of the ecumene, here city names and streets proliferate.

Nonetheless, this abundance of details is not the result of precise geographical knowledge of the region. The depiction appears chaotic and in some cases utterly wrong. The Mesopotamic fluvial system is far from being exact. The river Tigris, for example, has many incongruous characteristics. First, it gushes from a small mountain chain and then crosses another much longer one. Later, it flows into a neighboring river, the morphology of which is even more bewildering,44 and finally, after twisting with the latter, it flows into a circular inland body of water, named Palvdes (10C3). The number and location of cities and roads reflect a picture just as baffling. A few place names are written twice in two different positions of the map. This is the case with Sinjar, present on the map as Singara (10B5) and Sirgora (10C4, without a symbol), and Ain Sinu, on the map as Zagvrae (10B5) and Zogorra (10C4).45 As they doubled names, the mapmakers also doubled the relative routes (Singara-Hatris and Lacvs Beberaci-Singara). In addition to this confusion, one should mention the atypical position of the caption in question, which, due to the lack of space and the large amount of neighboring physical and urban elements, is vertical rather than horizontal.46

What is more perplexing about the entry TROGODITI PERSI is its content. The juxtaposition of the name “troglodytes” with the ethnonym Persians is unique. Late Antique Latin and Greek texts reflect a nuanced image of the Persians, who represented a sort of counterpart to the Roman world despite often being considered morally inferior.47 Accurate historical information, longstanding ethnic stereotypes, and literary metaphors and commonplaces interweave in the works of Late Antique writers, even in the writings of authors like Ammianus Marcellinus, Procopius, and Agathias, who either travelled to the eastern frontier or had (or claimed to have had) personal contact with Persians and access to Persian documents.48 Even if the depiction of Persians could vary according to author and political climate, Roman persons of letters shared a profound interest in this gens. The fourth-century historian Ammianus Mercellinus represents one of the most glaring examples of this fascination. In his historical work, known as the Res Gestae, he inserted a large number of excursuses which contain information primarily of a geographical and ethnographical nature. The section dedicated to the Persians, included in book 23 right before the account of Julian’s Persian expedition, is by far the longest.49 Since the chronology of this work (380s) is not very distant from the last redaction of the Tabula Peutingeriana, a brief analysis of Ammianus’ Persians highlights the extent to which the entry in question deviates from or converges with the opinions of his contemporaries.

Ammianus never explicitly defines the Persians as barbarians,50 yet his judgment of them cannot be considered positive. Century-old ethnographic stereotypes influenced the description of their physical features, temperament, and habits. In its desire to emulate more ancient and authoritative authors, such as Herodotus and Ptolemy, Ammianus’ digression resembles a display of erudition rather than a report of new information about Persian society. Although he seems well aware of the complexity and vastness of the Persian Empire, which is considered a patchwork of diverse peoples and disparate environments, Ammianus does not make any clear distinction among the subjects of the King of Kings when he lists the alleged virtues and vices of the Persians.51 Among their many moral flaws, he mentions their unrestrained lust (which explains why they have numerous concubines and as many wives as they can support), their effeminate posture, their vanity, and their cruelty. Ammianus’ remarks on king Sapor II (309–79) are everything but flattering: he is greedy, quick-tempered, rough, pompous, treacherous, and dishonest. Yet, Persians do not know pederasty, and they do not engage in obscene behavior, such as urinating in public. They are also extremely frugal when it comes to food and particularly disciplined on the battlefield.52

In other words, Ammianus stresses the Persians’ otherness, emphasizing their effeminacy, their licentiousness, and their cruelty, all typical traits of eastern barbarians according to Greco-Roman ethnography,53 but he also recognizes some praiseworthy aspects of their way of life, distancing himself in a few instances from the older historical tradition.54 To conclude, although the Persians share some characteristics with other barbarians and are often depicted in negative terms, they are a unique interlocutor for the Romans, an alius orbis55 representing another, although not equal, civilization. The judgment that shines through the pages of the Res Gestae seems to contradict the entry on the Tabula Peutingeriana, which puts the Persians unambiguously in the realm of the barbaricum.

However, the unusual connection of the Persians with the “troglodytes” appears less strange if one broadens the scope of the primary sources taken into consideration. While Ammianus attributes barbaric habits to Persians but never explicitly calls them barbarians, other Late Antique sources do define them as such. For example, a register of provincial, urban, and ethnic names dating to 314, the so-called Laterculus Veronensis,56 does not imply any difference between Persians and other barbaric groups. The ethnonym Persae is included in a list of gentes barbarae who spread under the authority of Roman emperors.57 For the author of this catalogue, there is no substantial difference between Persians and other barbaric groups, like Saxons, Vandals, and Goths.

One can recognize the same reasoning in a certain number of inscriptions dedicated to the emperor Julian and found in the eastern part of the empire.58 These inscriptions praise the emperor for his military, civic, and religious policies. Julian is celebrated as liberator orbis Romani, as restaurator templorum, and as recreator curiarum et rei publicae. Regarding Julian’s success over external enemies, the inscriptions contain the cognomina devictarum gentium Alamannicus, Francicus, Germanicus, and Sarmaticus.59 In a few cases, the emperor is hailed with more comprehensive victory titles, such as debellator omnium barbararum gentium, extinctor barbarorum or νικητὴς παντὸς ἔθνους βαρβαρικοῦ.60 In texts from the eastern provinces, the term barbari would likely indicate primarily the Persians, who represented the major threat in the area.61 This seems to be the most logical conclusion concerning at least two inscriptions from the Roman province of Phoenicia,62 where the text presents both cognomina, referring to individual groups of western barbarians, and the generic title extinctor barbarorum. The cognomina devictarum gentium allude to successes accomplished by Julian at the Rhine frontier in 355–58, while the pompous title extinctor barbarorum reflects the propaganda implemented by Julian and his supporters in the months preceding the Persian campaign.63

Although it does not contain the term barbari, an episode in Ammianus’ account of the Persian campaign represents one of the closest examples to the disparaging entry on the Tabula Peutingeriana. In front of an army that was increasingly demotivated and in need of supplies due to the effective scorched-earth policy of the Persians, Julian ordered some prisoners to be brought before the army and harangued his troops as follows: “Behold what those warlike spirits consider men, little ugly dirty goats; and creatures who, as many events have shown, throw away their arms and take to flight before they can come to blows.”64

The description refers to undernourished and unkempt prisoners, yet it repeats and amplifies negative stereotypes of Persians in general. The animal metaphor serves to highlight their physical repugnance and their cowardice on the battlefield. This tirade reflects the intolerant attitude towards everything that was not Roman, an attitude which formed an integral part of the political discourse supported and spread by Julian and his court.65

Libanius, a teacher of Greek rhetoric in Antioch who was a friend of and advisor to the emperor, was among the most prominent spokespersons of this anti-Persian rhetoric.66 In his orations, Persians are repeatedly and explicitly called barbarians.67 In Embassy to Julian, Libanius ascribes two quintessential barbarian vices to the Persians: the disdain for blood ties and the lack of mercy. Prone to violent outbursts, Persians act like wild beasts, while Julian is a Greek who rules over Greeks, and therefore follows a superior moral code of conduct.68 The Greek-Persian dichotomy follows the opposition human-inhuman. The political ideology supported and spread by Julian and his pagan collaborators tended to stress the Hellenic nature of Roman power.69 As stated more than once by Julian himself, Romans and Greeks belong to the same γένος: the Greeks civilized the Romans and the latter acquired, preserved, and spread the Greek religion and political institutions.70 In the political message of Julian and Libanius, the more the Romans resembled the Greeks, the more the Persians took the role of barbarians par excellence.

Although particularly evident in the works of Julian and his court, this attitude towards the Persians was not exclusive to their political and cultural discourse. Judgments of the Persians went hand in hand with the contemporary political situation. Since the rise of the Sasanian Empire in 224, Romans and Persians were in almost constant conflict.71 Mesopotamia, Syria, and Armenia were the main war zones. In a Roman world which looked on the Persians with renewed apprehension, the narrative of Alexander the Great enjoyed a period of revival. The Latin rendition of the Alexander Romance represents one pivotal example of this new interest in the figure of Alexander.72 This text, whose author is traditionally identified as Julius Valerius Alexander Polemius, consul in 338, contributed greatly to the diffusion and longevity of the myth of Alexander in medieval Europe. The Macedon and his deeds in the East represented a model for any Roman emperor who had to confront the Persian threat. If Julian was the most enthusiastic emulator of Alexander,73 other emperors aspired to follow in his footsteps.74 These are the premises on which a text known as the Itinerarium Alexandri (the latest possible date of which is 345) rests.75 Dedicated to Constantius II, this work exploits the myth of Alexander for contemporary political exigencies. Alexander’s expedition is presented both as an archetype and as an omen for the emperor, who had just started his campaign against the Persians. Significantly, the revival of the Alexander narrative also finds expression on the Tabula Peutingeriana. References to Alexander’s deeds play a central role in the map’s portrayal of the eastern lands. The campaigns of the Macedonian king are evoked through the numerous cities that bear his name (founded during or after Alexander’s reign),76 the mention of the Indian elephants,77 and especially two isolated symbols (the “altars of Alexander”), which, marking the limits of Alexander’s expeditions, define the edges of the inhabited world.78 Thus, it appears that the Alexander narrative enjoyed a period of renewed interest in Late Antiquity, a phenomenon that could be interpreted as closely linked to the contemporary political climate. The account of Alexander’s Persian campaign provided a story in which the Persians played the role of the main antagonist, who eventually succumbs, and thus the narrative served to reassure a Roman public worried about the aggressive Sasanian policy.

In light of the above, it is now possible to contextualize the entry TROGODITI PERSI on the Tabula Peutingeriana, which at first glance appears so bizarre. Contrary to the more nuanced judgment of influential historians, such as Ammianus and Procopius, the imperial discourse, influenced by the renewed popularity of the Alexander narrative, described the Persians in clear-cut negative terms. The Tabula Peutingeriana, or at least its last version, appears as the product of Roman imperial ideology. With Italy covering one-third of the map and Rome located in its center,79 the map represents the ecumene seen through the lens of Roman geography and political discourse. Moreover, the myth of Alexander, which offers a particularly disparaging image of the Persians, evidently informs the depiction of the East on the Tabula. To conclude, although the Tabula Peutingeriana represents the only instance in which the Persians are described as “troglodytes,” this legend can be interpreted as an extreme example of Late Antique anti-Persian rhetoric, which, fuelled by the political tensions at the time, repeated and adjusted themes of Alexander’s narrative and perpetuated the most derogatory stereotypes of the Persians.

The Case of the Persians Part II – South Arabia

The basis of the analysis in this article on South Arabia in the tenth century is the writings of al-Ḥasan b. Aḥmad b. Yaˁqūb al-Hamdānī (280–334 AH/894–945), a distinguished scholar, poet, and public figure. As one of Yemen’s minority groups, the Persians offer an example of al-Hamdānī’s adaptation of ethnic terminology, analyzed in consideration of its historical and ethnographic context. Applying ethnonyms in order to differentiate social categories is a universal strategy of othering. In the case of the Yemeni author, it becomes obvious that these strategies can only be understood in relation to simultaneous selfing processes, that they are primarily local, and that they are always contextual. An understanding of othering as the construction of an imagined other through the differentiation of this other from the self (often in a pejorative way) reveals a close link between this strategy and the concept of ethnicity itself, since processes of ethnic differentiation are generally based on constructions of precisely these kinds of dichotomies. The following terminological examination will clarify this.

Al-Hamdānī applies different ethnonyms to Persians in his writings: al-abnāˀ, al-furs, and al-ˁajam. Their etymological meaning and the context in which they are used in the sources are the basis of interpretation regarding the social implications of the terms and the author’s discursive strategies. Processes of selfing and othering concerning South Arabians and Persians are significantly shaped by tribal ideologies in al-Hamdānī’s account. However, his use of the different terms is fluid and cannot be clearly categorized. In spite of sectarian conflicts and continuous power struggles, particularly in the ninth and tenth centuries80 (when al-Hamdānī was writing), there is a remarkable element of continuity, most clearly expressed in the consistent use over long periods of time of tribal names and toponyms, which resist political ruptures and changes.81 This element of continuity must not be ignored in the study of tribal identities and social environments in the Yemeni highlands. The case of the Yemeni Persians further supports the argument. For many centuries, their main area of settlement remained the city of Sanaa and the surrounding region, and their most characteristic ethnonym al-abnāˀ appears consistently in the sources for about 600 years between the sixth and twelfth centuries.82

This main term for Persians (and also the term used most by al-Hamdānī), al-abnāˀ (“the sons”) in the abridged or abnāˀ al-furs (“the sons of the Persians”) in the complete form, clearly refers to the descendants of Persians, who came to Yemen at the end of the sixth century, when it fell under Sasanian rule.83 They were not regarded as “real Persians,” since they were born in Yemen and often had Yemeni mothers.84 Hence, the significance of the term is deeply rooted in the Yemeni local context and history. Al-Hamdānī mentions al-abnāˀ several times, particularly in his Kitāb Ṣifat jazīrat al-ˁarab (Geography of the Arabian Peninsula).85 Among them were prominent personalities, individual inhabitants of towns or villages, and larger groups of the population. The designation al-abnāˀ creates a terminological relation to the tribal population of Yemen. It has the same meaning as banū (“sons”), which is a term for members of a tribal group and can be part of the tribal name. Hence, the two terms abnāˀ and banū are equivalent designations with regard to meaning, yet they are distinct markers of social groups, which is highly interesting from the perspective of an analysis of the constructions of identity and interethnic relations. Both terms are part of group designations which imply kinship references. Through banū a connection to genealogy is expressed, which marks an important factor of tribal identities. However, not every tribal group shares such an identity-establishing genealogical record or is composed of a mixture of different genealogical backgrounds. Yet the addition of banū to the tribal name conveys the impression of having one shared genealogy.86

Al-furs appears either as additional part in the construction with abnāˀ, e.g. abnāˀ al-furs, or can otherwise be applied as an ethnonym by itself. Al-Hamdānī used al-furs partly synonymously with al-abnāˀ for people of Persian descent. The following example is a passage on Persians from the mining city of al-Raḍrāḍ, who came under attack and had to flee the town. It shows that al-furs is the term applied to descendants of Sasanian Persians but also of Persians who came to Yemen in later periods (under the Umayyads and the Abbasids). It seems they worked in the mine and therefore were called furs al-maˁdin (“Persians of the mine”). Some of them had a background in Sanaa, including houses [manāzil] and estates [ḍiyāˁ], to which they could return. Furthermore, those who returned to Sanaa are identified by names, which all contain banū, followed by a Persian word as the first syllable,87 and the same ending syllable [ōye], which was very common to Persian names even if probably pronounced differently88:

 

When Muḥammad b. Yuˁfir was killed and these qabāˀil89 fell into distress because of that, some of them acted unjustly against its inhabitants, killed among them and ransacked them. Who remained flew, and they were dispersed in the bilād [country]. A qawm [body of men/women] of them went to Ṣanˁāˀ who had a footing there from times of old and dwelling houses and property. Its inhabitants were all from al-Furs [the Persians], from those who came there during the jāhiliyya [pre-Islamic times], in the days of the Banū Umayya [Umayyads] and the Banū al-ˁAbbās [Abbasids]. They were called Furs al-maˁdin [Persians of the mine]. Who is in Ṣanˁāˀ of them are Banū Sardōye, Banū Mihrōye, Banū Zanjōye, Banū Bardōye, and Banū Jandōye.90

Al-furs is also a general designation for Persians beyond the Yemeni context, which refers to a territory, namely the bilād fāris (“Persia”). Fāris, “Persia,” was used in Achaemenid (559-330 BCE) and Sasanian (224-651 CE) times and designated both the Persians as an ethnic group and their homeland. In early Arabic sources, the term fārs/fāris was applied both in the narrow sense to the Persian province of Fars and in a wider sense to the whole Persian territory. As an ethnonym for Persians, al-furs was much more common than fāris.91 Al-furs can be opposed to other ethnonyms, such as al-ˁarab (“the Arabs”) or al-rūm (“the Byzantines”). Having the qualities of a typical ethnonym, al-furs cannot be combined with these alternative categories, since, when used in the same context, they are mutually exclusive.92 It is used in this sense by al-Hamdānī, for example in listings of people or lands, but since the focus of his writing is South Arabia, such lists are only marginal and little explanation of them is provided. One corresponding example is a passage from the Kitāb al-Jawharatayn al-ˁaṭīqatayn al-māˀiˁatayn min al-ṣafrāˀ wa-l-bayḍāˀ (The Book on the Two Noble Metals Gold and Silver) on the mining business in Yemen: “The merchants from among the Iraqis [al-ˁirāqīyīn], Persians [al-furs], Syrians [ash-shaˀmīyīn], and Egyptians [al-miṣrīyīn] carried away the silver of Yemen at that time, and they gained through it significant profit.”93 Here, Persians are listed with other agents active in the mining business. In this context, al-furs is in line with the other foreign categories, and there is no indication of any closer relation to Yemen. On the contrary, there is some sign of a tie to the foreign lands to which the silver was “carried away” by the merchants.

Another designation that can be applied to Persians is al-ˁajam (pl. aˁājim). Al-Hamdānī mentions it more rarely than al-abnāˀ and al-furs. Where it appears in the text, it is sometimes not clear whether it actually refers to Persians or to non-Arabs. This ambivalence is caused by the historical use of the term. ˁAjam has its etymological root in ˁujma, “impure speech,” and is opposed to faṣāḥa, “highly eloquent, clear speech.” Even pre-Islamic poetry drew a distinction between al-ˁarab and al-ˁajam. In the context of the Islamic conquests, it was used in order to distinguish between “Arabized” populations and “pure” or “real” Arabs.94 The etymology and semantic evolution of this collective term are comparable to those of the Greek term βάρβαροι, and it was primarily associated with the neighboring Persians. The affective value attributed to the word was at times inspired by claims of Arab superiority due to their more civilized and refined culture.95 In Yemen of the tenth century, al-ˁajam could have functioned as general designation for non-Arabs, could have meant non-Arabic speakers, or could have been an ethnic designation for Persians. Al-Hamdānī, for example, writes about abnāˀ aˁjam,96 which can be translated as “offspring of non-Arab descent” or “offspring of Persian descent.” In later sources, such as Ibn al-Mujāwir’s Taˀrīkh al-Mustabṣir, al-ˁajam was explicitly used for Persians. The Shuˁūbīya movement in the time of the ‘Abbasids97 questioned Arab superiority and strove to revalue the role of the aˁājim, which mostly but not exclusively meant the Persians. Fostered by these developments, ˁajam as the initially pejorative identification of others by Arabs became a neutral term of ethnic differentiation. With its novel quality of an ethnic group designation for Persians, it was eventually used as self-ascription.98 Moreover, the term could also be used to denote a territory, i.e. bilād al-aˁājim (“non-Arab lands”).99 Thus, ˁajam in its various forms is the most unspecific of the terms in question. It stresses distinctiveness without qualifying or defining it. It can be assumed that the Yemeni readers were able to discern, at least in some matters, whether al-ˁajam meant Persians or non-Arabs, but al-Hamdānī’s intention in using this term might have been less to identify the other and more to evoke a sense of it.

In order to explore ethnonyms for Persians in South Arabia in a later medieval period, this case study draws on Ibn al-Mujāwirs Taˀrīkh al-Mustabṣir (The Historiography of the Sharp-Sighted), a travelogue from the early thirteenth century. The author, to all appearances a native of Khorasan, Persia and native Persian-speaker, visited South Arabia at least three times between 1220 and 1230. He shows great interest in the history and topography of the Arabian Peninsula, but mainly focuses on the topic of trade and commerce, which leads to the assumption that he was most likely a businessman himself.100 In a copious style, he combines his own observations and those of informants and transmitters with an abundance of storytelling, including local myths and legends as well as Quranic themes and his own dreams, creating a rich mix of genres. Thus, the analysis of ethnonyms in this source focuses on their roles and meanings in these different narrative units and also their etymology and historical context.

In Taˀrīkh al-Mustabṣir, the abovementioned terms al-furs and al-ˁajam are used synonymously and are equally eligible to designate people of Persian origin or descent, meaning either Persians who came to Yemen from Persia or their Yemeni offspring. It is clear that the term ˁajam refers to no other ethnicity than Persian. ˁAjam as used by Ibn al-Mujāwir has no negative connotation whatsoever and bears no reference to its Greek/Arabian etymology of “impure speech.” The author even uses the term al-ˁAjamīya to refer to the Persian language when quoting a Persian saying given by his contemporary Yemeni Persian-speakers. Although furs and ˁajam are used interchangeably, the term ˁajam is used slightly more often in the text. When in opposition to other typical ethnonyms, especially ˁarab, the term ˁajam is preferred, maybe because of its phonetic similarities or because al-ˁarab wa-l-ˁajam had been coined as a pair of opposites since pre-Islamic times. The following quotation shows how ˁajam qualifies as a term that clearly designates Persians as an ethnic group in distinction to others:

 

They are a people descended from Ham, son of Noah – peace be upon him. Moreover they are not Arabs [ˁarab], Persians [ˁajam], Indians [hind], Abyssinians [ḥabash], Turks [turk] nor Nabateans [nabaṭ], but they have a language all of their own which is used [only] among themselves.101

In this passage, the author recounts one of his dreams about a mystic valley and its inhabitants near the city of Medina. In this case, it is obvious that “Persian” serves as an ethnic category that is dissociated from any distinct historical timeframe.

Whenever the frequently used phrase “it was built by the Persians” or its variation, “a construction of the Persians,” is used, the question arises how an approximate historical period can be determined. Ibn al-Mujāwir provides his readership with a hodological rather than a chronologically organized narrative, meaning that he structures his writing according to the stops on his itinerary (towns, cities, historical sites). He shows a pronounced tendency to attribute the erection of impressive building structures such as fortifications and mosques and even the founding of whole towns to the Persians, whether these edifices or settlements still existed, were in ruins, or were gone altogether. Flipping this causality around, he interprets the remains of constructions as indications of Persian presence in major cities and various other towns in earlier history. If at all, he gives only vague time specifications, e.g. “when the Pharaonic rule came to an end,” “in the days of the Persians,” and “under Persian rule.” One can either try to reconstruct at least a tentative timeframe from the context and compare it with other historical records to determine whether the building in question was or could have been of Persian making, or one can use additional information given on the building materials, location (e.g. the center or the periphery of the town), and other architectural features to determine Persian workmanship.102 Particularly in the case of Aden, it seems most likely that expressions like “the days of the Persians” refer to the period of Sasanian rule in the sixth century. Roxani Margariti notes that “Al-Marzūqī103 conveys the tradition that Aden always came under the jurisdiction of Yemen’s rulers.”104 On some occasions, Ibn al-Mujāwir supports his statements by saying that the information was revealed to him in a dream, hence it was a divinely inspired vision (manām or ruˀyā), which can be understood as strong proof.105 A legend like that of Alexander the Great, a mythical figure who also appears in the Quran, might also serve as evidence. In spite of the fact or, perhaps, precisely because of the fact that such narratives defy clear historical substantiation, they attain a strong effect. The following may serve as a prime example:

 

When Dhū l-Qarnayn [Alexander the Great] released the sea from Jabal Bāb al-Mandab and it flowed out, all the area around Aden dried up. […] When the Persian rulers [mulūk al-ˁajam] took over Aden they saw this exposed area and were afraid for the town, that someone coming to conquer might lay siege to it. Then they made an opening on the side near to Jabal ˁImrān and released the sea over it. The sea poured forth, descending until it drowned the whole exposed area around Aden. Aden became an island. […] The new-made sea was called Buḥayrat al-Aˁājim and was known by their name for all times.106

Here, the plural form aˁājim is part of the hydronym buḥayrat al-aˁājim. Ibn al-Mujāwir never uses the word aˁājim as a term to designate the Persians. It occurs only twice in the text, the second time in a piece of Arabic poetry which he quotes. Mulūk al-ˁajam might refer to the Sasanian rulers, but the fact that this story connects to the legend of Dhū l-Qarnayn creates a certain level of ambiguity. Such ambiguities are characteristic of Ibn al-Mujāwir’s writing and do not necessarily undermine the author’s discourse. If anything, they create narrative tension and draw more attention to what seems to be the author’s intention, namely to point out the momentousness of Persian influence in medieval Yemen.

One group of Persians which is datable and clearly distinguishable from other Persians throughout the text are al-furs min ahl Sīrāf, “the Persians of the people of Siraf.” The ancient city of Siraf, situated on the Iranian coast of the Persian Gulf, was a seaport and early Islamic trade center. In 997, it was left in ruins by an earthquake which lasted for seven days. The people of Siraf, whose merchants had already been traveling back and forth to the Red Sea, then immigrated to the coastal regions in the area.107 Ibn al-Mujāwir introduces the ahl Sīrāf early in his account when dealing with the history of the seaport Jeddah.108 A group of contemporary Yemeni Persians tells him the story of their Sirafi ancestors who fortified the city by enclosing it with a massive wall. They then dug a huge moat around it so that the seawater would pour into it and run around the town until it flowed back into the sea. Thus the city of Jeddah resembled an island amidst the sea. The incredible number of a thousand reservoirs, built to guarantee a secure supply of drinking water, adds a fantastic element to the story. After approximately 80 years of prosperous community life, the Persians were forced out of Jeddah by the Arabs, and they immigrated to other islands and coastal cities in the region yet again.

Another tale which features the ahl Sīrāf speaks to their pride and wealth as merchants.109 The same story appears in Ibn al-Baṭṭūṭa’s Riḥla, who visited Aden about a hundred years later, probably around 1330.110 In Ibn al-Mujāwir’s account the protagonists are the Sirafi Persians. Two slave-boys of two Sirafi merchants are sent to the market to bid for fish. The slaves start bidding for the only fish available until the price exceeds 1,000 Dinars and one of them buys it. When he brings home the fish, his master is so pleased with him that he sets him free and provides him with 1,000 Dinars sustenance. The other slave who returns to his master empty-handed is severely punished. In Ibn Baṭṭūṭa’s version the fish is a ram, but the masters who send their slave-boys are not associated with any particular ethnic group, but rather with the social group of Adeni traders in general.111

As to the etymology of the name Aden, Ibn al-Mujāwir states that it was derived from the word maˁdin, more specifically from maˁdin al-ḥadīd, “the iron mine,” and that it was called ākhuri sangīn “an empty, or rather, stony cratch” by the Persians.112 This is clearly a reference to the Persian mining activities in Yemen and a further example of how the author uses different tools to point out the strong impact of Persian presence in medieval Yemen.

Despite the different etymologies of al-furs, al-ˁajam and al-abnāˀ, and although there are some tendencies in al-Hamdānī’s texts, which at first glance suggest preferences for one or the other term depending on the context, these South Arabian sources from the tenth century show that no clear-cut distinctions between the three terms can be made. The range between al-abnāˀ, al-furs, and al-ˁajam varies in terms of their othering potential. The relatedness of the terms abnāˀ and banū can be interpreted as minor differentiation and could even be read as a strategy of selfing, e.g. al-abnāˀ refers to the Yemeni Persians, understood as part of the author’s own society, in contrast to al-furs, which refers to the Persian Persians, not understood as part of Yemeni society. Of course, the differentiation between abnāˀ and banū, or Persians and Yemeni tribesmen respectively, continues. Otherwise the ethnonym would not make sense. Consequently, the three ethnonyms for Persians in Yemen, which al-Hamdānī uses, combine different levels of selfing and othering. Yet the flexibility of how they are used underlines the fluid character of ethnic categorizations in medieval South Arabia. Genealogy and descent were major factors of tribal belonging and ethnic naming, as the case of the abnāˀ shows. What becomes evident is that in the case of the abnāˀ, the construction of identity for the (Arab) self and the (Persian) other follows patterns of tribal belonging and genealogical descent. The abnāˀ are addressed as sons/offspring of an ancestral group or referential figure, like Yemeni tribesmen, but this ancestral reference is an ethnic other: the furs (abnāˀ al-furs). In al-Hamdānī’s writings, ˁajam and its variant forms bear, according to its etymology, the highest othering potential. Yet the affective dimension of the term gradually lost its meaning of “impurity” in its practical applications over time. Also, the term became increasingly limited to designating something or someone as “Persian,” rather than referring to a non-Arab or non-Arabic other. Ibn al-Mujāwir, in all probability a native Persian and an author of the early thirteenth century, also uses the terms furs and ˁajam to designate Persians in the Yemeni context, and a third term which refers to a certain group of immigrants, al-furs min ahl Sīrāf. Furs and ˁajam are used synonymously and are predominantly to be understood as ethnic terms which identify people of Persian descent. Both furs and ˁajam convey no additional information as to social status or the historicity of a group, unless they are combined with other compounds which indicate either rulership or geographical origin, e.g. mulūk al-ˁajam or al-furs min ahl Sīrāf. Ibn al-Mujāwir’s writing is characterized by fluid transitions between the historical report, mythical narratives, fantastic stories, and genuine observations, a style that creates ambiguity, which can be interpreted as part of the author’s narrative strategy. The text consistently highlights the Persians’ presence in South Arabia, and building structures serve as the main indicator in the emerging discourse of Persian self-authentication. The ethnonyms al- furs, al-ˁajam and (al-furs min) ahl Sīrāf are all suitable to praise the outstanding accomplishments of Persians and thereby portray them as an ethnic group who is more advanced in comparison to others. Thus, all three terms have the same selfing potential. This is also where it becomes most evident that over time the term ˁajam not only lost its original pejorative meaning, but eventually acquired a positive affective value, being used as a Persian self-ascription.

Conclusion

This comparative study shows that ethnonyms function as conceptual tools, which authors can strategically use in their narratives. Drawing on Latin, Greek, and Arabic source material, we presented the particular case of the Persians as an illustration that ethnonyms are dynamic and adaptable, that they shape processes of selfing and othering, and that they support ideologies and political agendas. Ethnonyms and, more generally, ethnic terminology can serve to accentuate or reduce the hiatus between the self and the other. The case study shows that constructions of Persians as the other in interethnic relations varied greatly over the course of time.

The legends that refer to the Persians on the Tabula Peutingeriana are a mixture of erudite citations, ill-informed guesses, and ethnographic commonplaces. While Late Antique Greek and Latin authors have conveyed a nuanced view of the Persians and their empire, the inscription TROGODITI PERSI on the Tabula represents a vivid example of ethnic polemic in the service of Roman imperial propaganda.

TROGODITI PERSI on the Tabula Peutingeriana portrays the Persians as an extreme opposite to the Roman (self-)understanding of (Roman) “civilization.” Around the time of the advent of Islam, Arabs developed a comparable notion of “non-Arabs” (al-ˁajam or al-aˁājim), from which they marked themselves off. In the case of the Roman map, the point of reference for distinction was the dwelling place and way of life, whereas the Arabs referred to language. For them, Arabic was the divine language, and it was closely linked to the holy script of the Quran, which distinguished the Arabs from all non-Muslims and non-Arabs. Persians were among the first non-Arabic speakers whom the Arabic-speaking Muslims conquered in the seventh century. Until the tenth and thirteenth centuries, ˁajam was more and more closely associated with (and even adapted by) the Persians, and less closely associated with non-Arabs in general, or the inability to speak Arabic properly. In Latin sources, the terminological designations for Persians did not change significantly between Antiquity and the Middle Ages. However, perceptions of the Persae varied between notions of them being “barbarous” in the fullest sense to them being “civilized,” or even similar and comparable to the Romans.

In the South Arabian context of the tenth century, the local term al-abnāˀ created a zone of transition between the Yemeni self and notions of the Persian other. Al-Abnāˀ could be combined or exchanged with the transregional terms al-furs and al-ˁajam to appropriate new meanings. In this interplay of local and transregional ethnonyms, it was possible either to enhance the role of the Persian minority as an integral part of the Yemeni society or to express a stronger sense of its otherness and separation. In the early thirteenth century, the Persian author Ibn al-Mujāwir used the terms al-furs and al-ˁajam to elevate the Persians as a civilization in the context of South Arabian history. A strong element of storytelling, together with references to elements of construction as evidence of civilizational accomplishments, built the framework through which he engaged in a discourse of Persian self-authentication.

Acknowledgments

The research for this article was funded by a DOC-team-fellowship provided by the Austrian Academy of Sciences. We would like to thank Johann Heiss, Walter Pohl, Stephan Prochazka, Danuta Shanzer, and Ekkehard Weber for their support and advice during the preparation of this article and the preceding research phase.

Bibliography

Primary Sources

 

Latin and Greek

Ammianus Marcellinus. Res Gestae. Edited by W. Seyfarth. Leipzig: Teubner, 1978.

Dio Cassius. Roman History. Volume 9. Books 71–80. Translated by Earnest Cary. With Herbert B. Foster. Loeb Classical Library. Cambridge, MA: Harvard University Press, 1927.

Digest. Edited by T. Mommsen, and P. Krüger. Berlin: Wiedmann, 1889.

Eusebius of Caesarea. Life of Constantine. Edited by F. Winkelmann. Berlin: Akademie Verlag, 1975.

Geographi Latini Minores. Edited by A. Riese. Heilbronn: Henningfer Brudern, 1878.

Herodian. History of the Empire, Volume 2. Books 5–8. Translated by C. R. Whittaker. Loeb Classical Library. Cambridge, MA: Harvard University Press, 1970.

Herodotus, Historiae. Edited by H. B. Rosé. Leipzig: Teubner, 1987.

Julian. The Caesars; Hymn to King Helios. Edited by C. Lacombrade. Paris: Les Belles Lettres, 1964.

Libanius. Opera. Edited by R. Foerster. Leipzig: Teubner, 1903–1927.

Origo Gentis Langobardorum. Edited by G. Waitz. Hannover: n.p., 1878.

Panegyrici Latini. Edited by R. Mynors. Oxford: Clarendon Press, 1964.

Paul the Deacon. Historia Langobardorum. Edited by G. Waitz. Hannover: n.p., 1878.

Paschoud, F. Zosime, Histoire Nouvelle. 2. vols. Paris: Les Belles Lettres, 1971.

Synesius. On Imperial Rule. Edited and translated by A. Fitzgerald. Oxford: n.p., 1927.

 

Arabic

Ath-Thaqafī, Sulaymān b. Yaḥyā. Sīrat al-ˀImām ˀAḥmad b. Sulaymān, 532–566 H (Taˀrīkh al-Jazīrah al-ˁArabīya), first edition. ˁAyn for Human and Social Studies, 2002.

Ibn Baṭṭūṭa. Riḥlat Ibn Baṭṭūṭa. Beirut: Dār Ṣādir, Dār Bayrūt, 1964. S252

Kramers, J. H., ed. Opus geographicum: secundum textum et imagines codicis Constantinopolitani conservati in Bibliotheca Antiqui Palatii N°3346 cui titulus est “Liber imagines terrae” [Kitāb ṣūrat al-ˀarḍ]. Leiden: Brill, 1873 [reprint 1967].

Löfgren, O., ed. Arabische Texte zur Kenntnis der Stadt Aden im Mittelalter: Abu Maẖramas Adengeschichte nebst einschlägigen Abschnitten aus den Werken von Ibn al-Muǧāwir, al-Ǧanadī und al-Ahdal. Uppsala: Almqvist & Wiksells Boktryckeri A. B., 1936.

Löfgren, O., ed. Ibn al- Mujāwir: Taˀrīkh al-Mustabṣir, Teil 1. Leiden: Brill, 1951.

Löfgren, O., ed. Ibn al- Mujāwir: Taˀrīkh al-Mustabṣir, Teil 2. Leiden: Brill, 1954.

Müller, D. H. al-Hamdānī’s Geographie der arabischen Halbinsel: nach den Handschriften von Berlin, Constantinopel, London, Paris und Strassburg zum ersten Male herausgegeben [Kitāb Ṣifat jazīrat al-ʿarab]. Leiden: Brill, 1968.

Ptolemy, Tetrabiblos, with an English translation by F. E. Robbins. Cambridge, MA: Harvard University Press, 2014. Accessed July 27, 2017. Loeb Classical Library, https://0-www-loebclassics-com.wam.leeds.ac.uk/view/LCL435/1940/volume.xml.

Toll, C. Kitāb al-Ǧawharatain al-ˁaṭīqatain al-māˀiˁatain aṣ-ṣafrāˀ wa ’l-baiḍāˀ. Die beiden Edelmetalle Gold und Silber, by al-Ḥasan Ibn-Aḥmad al-Hamdānī. Uppsala: Almqvist & Wiksell, 1968.

 

Secondary Sources

Al-Azmeh, Aziz: “Barbarians in Arab Eyes.” Past & Present 134 [1992]: 3–18. Oxford: Oxford University Press: 1992.

Albu, E. The Medieval Peutinger Map: Imperial Roman Revival in a German Empire. Cambridge–New York: Cambridge University Press, 2014.

Benoist, B. “Identité du Prince et discours impérial: Le cas de Julien.” Antiquité Tardive 17 (2010): 109–17.

Bernard, Gwladys. “Les prétendues invasions maures en Hispanie sous le règne de Marc Aurèle: essai de synthèse.” Pallas 79 (2009): 357–75.

Bianchetti, S., M. Cataudella, and H.-J. Gehrke. Brill’s Companion to Ancient Geography: The Inhabited World in Greek and Roman Tradition. Leiden: Brill, 2015.

Blockley, R. C. East Roman Foreign Policy: Formation and Conduct from Diocletian to Anastasius. Leeds: Caims, 1992.

Börm, H. Prokop und die Perser: Untersuchungen zu den römisch-sasanidischen Kontakten in der ausgehenden Spätantike. Stuttgart: Steiner, 2007.

Bowersock, G. Julian the Apostate. Cambridge, MA: Harvard University Press, 1997.

Braun, Christiane. “Untersuchungen zum XI. Segment der Tabula Peutingeriana anhand der Route Persepolis – Ekbatana – Hecantopolis – Propasta – Antiochia.” Orbis Terrarum 14 (2016): 11–32.

Cameron, A. “Agathias on the Sassanians.” Dumbarton Oaks Papers 23–24 (1969/1970): 67–183.

Cardona, G. R. Nomi propri e nomi di popoli: una prospettiva etnolinguistica, Urbino: Università di Urbino (Centro Internazionale di Semiotica e di Linguistica), 1982.

Caltabiano, M. “La comunità degli Elleni: cultura e potere alla corte dell’imperatore Giuliano.” Antiquité Tardive 17 (2009): 137–49.

Chauvot, A. “Parthes et Perses dans les sources du IVe siècle.” In Institutions, société et vie politique dans l’Empire romain au IVe siècle ap. J.-C. Actes de la table ronde autour de l’œuvre d’André Chastagnol (Paris, 20–21 janvier 1989), edited by M. Christol, S. Demougin, and Y. Duval, 115–25. Rome: École Française de Rome, 1992.

Chauvot, A. Opinions romaines face aux Barbares au IVe siècle ap. J.-C. Paris: De Boccard, 1998.

Conti, S. “Un aspetto della propaganda imperiale tardo-antica: la titolatura di Giuliano nelle fonti letterarie ed epigrafiche.” Koinonia 30–31 (2006/07): 29–44.

Conti, S. Die Inschriften Kaiser Julians. Stuttgart: Steiner, 2004.

Cracco Ruggini, L. “Pregiudizi razziali, ostilità politica e culturale, intolleranza religiosa nell’impero Romano.” Athenaeum 46 (1968): 139–52.

Crone, Patricia. “The ‘Abbāsid Abnā’ and Sāsānid Cavalrymen.” Journal of the Royal Asiatic Society, Third Series 8, no. 1 (1998): 1–19.

Dalché, P. G. “La trasmissione Medievale e Rinascimentale della Tabula Peutingeriana.” In Tabula Peutingeriana: Le antiche vie del mondo, edited by F. Pontera, 43–53. Florence: Leo S. Olschki, 2003.

Dalché, P. G. “L’enseignement de la géographie dans l’antiquité tardive.” Klio 96 (2014): 144–82.

“Dar al-Harb.” The Oxford Dictionary of Islam. Accessed August 30, 2017. http://www.oxfordislamicstudies.com/article/opr/t125/e490#.

Dauge, Y. A. Le Barbare: Recherches sur la conception romain de la barbarie et de la civilisation. Brussels: Latomus, 1981.

Davenport, C. “Imperial Ideology and Commemorative Culture in the Eastern Roman Empire 284–450 CE.” In Byzantium, Its Neighbours and Its Cultures, edited by D. Dzino, and K. Parry, 45–70. Brisbane: Australian Association for Byzantine Studies, 2014.

Den Boeft, J. Philological and Historical Commentary on Ammianus Marcellinus: XXIV. Leiden: Brill, 2002.

Desanges, J. “Fluminenses.” In Encyclopédie berbère. Vol. 19: Filage – Gastel, edited by Gabriel Camps, 2862. Aix-en-Provence: Edisud, 1998. Accessed December 2, 2017. http://encyclopedieberbere.revues.org/1946.

Dietz, K. “Kaiser Julian in Phönizien.” Chiron 30 (2000): 807–55.

Dodgeon, M. l. H., G. Greatrex, and Samuel N. C. Lieu. The Roman Eastern Frontier and the Persian Wars (Part I, 226–363 AD). London: Routledge, 2002.

Dodgeon, M. H., G. Greatrex, and Samuel. N. C. Lieu. The Roman Eastern Frontier and the Persian Wars (Part II, 363–630 AD). London: Routledge, 2002.

Dresch, P. Tribes, Government, and History in Yemen. Oxford: Clarendon Press, 1989.

Drijvers, J. W. “Ammianus Marcellinus’ Image of Sasanian Society.” In Ērān ud Anērān: Studien zu den Beziehungen zwischen dem Sasanidenreich und der Mittelmeerwelt. Beiträge des Internationalen Colloquiums in Eutin, 8.–9. Juni 2000, edited by J. Wiesehöfer, and P. Huyse, 45–69. Stuttgart: Steiner, 2006.

Drijvers, J. W. “A Roman Image of the “Barbarian” Sasanians.” In Romans, Barbarians and the Transformation of the Roman World, edited by R. W. Mathisen, and D. Shanzer, 67–76. Surrey: Ashgate, 2011.

Drijvers, J. W. “Rome and the Sasanid Empire: Confrontation and Coexistence.” In A Companion to Late Antiquity, edited by P. Rousseau, 441–54. Malden–Oxford: Wiley-Blackwell 2009.

Drijvers, J. W. “Ammianus Marcellinus’ Image of Arsaces and Early Parthian History.” In The Late Roman World and its Historian. Interpreting Ammianus Marcellinus, edited by J.W. Drijvers, and D. Hunt, 193–206. London–New York: Routledge, 1999.

Dunlop, D. M., “al-Balkhī.” The Encyclopaedia of Islam, second Edition, edited by P. Bearman, Th. Bianquis, C.E. Bosworth, E. van Donzel, and W.P. Heinrichs. Leiden: Brill, 2012. Accessed December 01, 2017. http://dx.doi.org/10.1163/1573-3912_islam_SIM_1157.

Eck, W. “Zur Neulesung der Iulian-Inschrift von Ma’ayan Barukh.” Chiron 30 (2000): 857–59.

Enderwitz, S. “al-Shuˁūbiyya.” The Encyclopaedia of Islam, Second Edition, edited by P. Bearman, Th. Bianquis, C.E. Bosworth, E. van Donzel, and W.P. Heinrichs. Leiden: Brill, 2012. Accessed March 14, 2017. http://dx.doi.org/10.1163/1573-3912_islam_SIM_6997.

Fahd, T., and H. Daiber. “Ruʾyā.” The Encyclopaedia of Islam, Second Edition, edited by P. Bearman, Th. Bianquis, C.E. Bosworth, E. van Donzel, and W.P. Heinrichs, Leiden: Brill, 2012. Accessed March 20, 2017. http://dx.doi.org/10.1163/1573-3912_islam_COM_0945.

Ferraco, F. Ammiano geografo: La digressione sulla Persia (23.6). Naples: Loffredo, 2004.

Ferreira, A. I. M. “El papiro 229 de la British Library: transcripción, traducción y estudio de un documento de compraventa.” Faventia 32–33 (2010/2011): 93–111.

Fowden, E. K. “Constantine and the Peoples of the Eastern Frontier.” In The Cambridge Companion to the Age of Constantine, edited by N. Lenski, 377–98. Cambridge: Cambridge University Press, 2012.

Fowden, G. “The Last Days of Constantine: Oppositional Versions and Their Influence.” The Journal of Roman Studies 84 (1994): 146–70.

Fowden, G. Empire to Commonwealth: Consequences of Monotheism in Late Antiquity. Princeton: Princeton University Press, 1994.

Gabrieli, F. “ˁAdjam.” The Encyclopaedia of Islam, Second Edition, edited by P. Bearman, Th. Bianquis, C.E. Bosworth, E. van Donzel, and W.P. Heinrichs. Leiden: Brill, 2012. Accessed March 10, 2017. http://dx.doi.org/10.1163/1573-3912_islam_SIM_0322.

Galichian, R. Countries South of the Caucasus in Medieval Maps: Armenia, Georgia and Azerbaijan. Yerevan–London: Artinfo Art Books/Gomidas Institute, 2007.

Gingrich, A. “Comparative Methods in Socio-Cultural Anthropology Today.” In The Sage Handbook of Social Anthropology, vol. 2, edited by Richard Fardon, Olivia Harris, Trevor H. J. Marchand, Mark Nuttal, Cris Shore, Veronica Strang, and Richard A. Wilson, 201–15. London: Sage, 2012.

Gingrich, A. “Multiple Histories: Three Journeys through Academic Records, Medieval Yemen, and Current Anthropology’s Encounters with the Past.” History and Anthropology, 2014. Accessed July 9, 2014. http://dx.doi.org/10.1080/02757206.2014.933102.

Guzmán Armario, F. J. “Ammianus adversus externae gentes: la geografía del Barbaricum en Amiano Marcelino.” Espacio, Tiempo y Forma, Serie II, Historia Antigua 12 (1999): 217–22.

Heiss, Johann. Unpublished Translation of Kitāb al-Ǧawharatain al-ˁaṭīqatain al-māˀiˁatain aṣ-ṣafrāˀ wa ’l-baiḍāˀ [al-Ḥasan Ibn-Aḥmad al-Hamdānī], by Christopher Toll, Uppsala: Almqvist & Wiksell, 1968 [2014].

Heiss, J. “Tribale Selbstorganisation und Konfliktregelung: Der Norden des Jemen zur Zeit des ersten Imams (10. Jahrhundert).” Diss., Universität Wien, 1998.

Isaac, B. The Invention of Racism in Classical Antiquity. Princeton: Princeton University Press, 2004.

James, B. “Arab Ethnonyms (ˁAjam, ˁArab, Badū and Turk): The Kurdish Case as a Paradigm for Thinking about Differences in the Middle Ages.” Iranian Studies 47, no. 5 (2014): 683–712

Johnstone, M. Jibbāli Lexicon. School of Oriental and African Studies. London: Oxford University Press, 1981.

Kaldellis, A. Procopius of Caesarea: Tyranny, History, and Philosophy at the End of Antiquity. Philadelphia: University of Pennsylvania Press, 2004.

Klein, R. “Laterculus Veronensis.” Lexikon des Mittelalters 5 (1999): 1745–46.

Kolb, F. Herrscherideologie in der Spätantike. Berlin: De Gruyter, 2001.

Lane Fox, R. J. “The Itinerary of Alexander: Constantius to Julian.” The Classical Quarterly 47, no. 1 (1997): 239–52.

Leveau, P. “L’aile II des Thraces, la tribu des Mazices et les praefecti gentis en Afrique du Nord.” Antiquités africaines 7 (1973): 153–91.

Lewis, B. “al-Abnāˀ.” The Encyclopaedia of Islam, Second Edition, edited by P. Bearman, Th. Bianquis, C.E. Bosworth, E. van Donzel, and W.P. Heinrichs. Leiden: Brill, 2012. Accessed September 4, 2013. http://referenceworks.brillonline.com/entries/encyclopaedia-of- islam-2/al-abna-COM_0008.

Maenchen-Helfen, J. O. The World of the Huns: Studies in Their History and Culture. Los Angeles: University of California Press, 1973.

Magini, M. “In viaggio lungo le strade della Tabula Peutingeriana.” In Tabula Peutingeriana: Le antiche vie del mondo, edited by F. Pontera, 7–15. Florence: Leo S. Olschki, 2003.

Malone, S. J. “Legio XX Valeria Victrix: A Prosopographical and Historical Study.” PhD diss., University of Nottingham, 2005.

Margariti, R. E. “An Ocean of Islands: Islands, Insularity, and Historiography of the Indian Ocean.” In The Sea: Thalassography and Historiography, edited by Peter N. Miller, 198–229. Ann Arbor: University of Michigan Press, 2013.

Margariti, R. E. Aden and the Indian Ocean Trade: 150 Years in the Life of a Medieval Arabian Port. Chapel Hill: The University of North Carolina Press, 2012.

Mathisen, R. “Natio, Gens, Provincialis and Civis: Geographical Terminology and Personal Identity in Late Antiquity.” In Shifting Genres in Late Antiquity, edited by G. Greatrex and H. Elton, 277–86. Farnham–Burlington, VT: Ashgate, 2015.

Matthews, J. F. The Roman Empire of Ammianus Marcellinus. Baltimore: The Johns Hopkins University Press, 1989.

McDonough, S. “Were the Sasanians Barbarians?” In Romans, Barbarians and the Transformation of the Roman World, edited by R. W. Mathisen, and D. Shanzer, 55–65. Surrey: Ashgate, 2011.

Migliorati, G. Iscrizioni per la ricostruzione storica dell’impero romano da Marco Aurelio a Commodo. Milan: EDUCatt, 2011.

Miller, K. Itineraria Romana: Römische Reisewege an der Hand der Tabula Peutingeriana. Stuttgart: Strecker und Schröder, 1916.

Miller, K. Die Peutingersche Tafel. Reprint. Stuttgart: Brockhaus, 1962.

Miquel, A., “Ibn Baṭṭūṭa.” The Encyclopaedia of Islam, Second Edition, edited by P. Bearman, Th. Bianquis, C.E. Bosworth, E. van Donzel, and W.P. Heinrichs. Leiden: Brill, 2012. Accessed August 29, 2017. http://dx.doi.org/10.1163/1573-3912_islam_SIM_3113.

Miquel, A., “Ibn Ḥawqal.” The Encyclopaedia of Islam, Second Edition, edited by P. Bearman, Th. Bianquis, C.E. Bosworth, E. van Donzel, and W.P. Heinrichs. Leiden: Brill, 2012. Accessed November 29, 2017. http://dx.doi.org/10.1163/1573-3912_islam_SIM_3193.

Mirza, S. “Dreaming the Truth in the Sīra of Ibn Hishām.” In Dreams and Visions in Islamic Societies, edited by Felek Özgen and Alexander D. Knysh, 15–30. New York: Suny Press, 2012.

Mittenhuber, F. Text- und Kartentradition in der Geographie des Klaudios Ptolemaios: eine Geschichte der Kartenüberlieferung Original bis in di Renaissance. Bern: Bern Studies in History and Philosophy, 2009.

Modéran, Y. Les Maures et l’Afrique romaine, IVe–VIIe s. Rome: Bibliothèque des Écoles françaises d’Athènes et de Rome, 2003.

Monda, S. La Cosmographia di Giulio Onorio: Un exceptum scolastico tardo-antico. Rome: Aracne, 2008.

Morley, C. “Beyond the Digression: Ammianus Marcellinus on the Persians.” Journal of Ancient History and Archeology 3, no. 4 (2016): 10–25.

Negev, A. “The Inscription of the Emperor Julian at Ma‘ayan Barukh.” Israel Exploration Journal 19, no. 3 (1969): 170–73.

Neuwirth, A. Introduction. In Myths, Historical Archetypes and Symbolic Figures in Arabic Literature: Towards a New Hermeneutic Approach, edited by Angelika Neuwirth, Birgit Embaló, Sebastian Günther, and Maher Jarrar, ix–xxi. Beirut: Steiner, 1999.

Oates, D., and J. Oates. “Ain Sinu: A Roman Frontier Post in Northern Iraq.” Iraq 21, no. 2 (1959): 207–42.

Palermo, R. “Settlement Patterns and Road Network in Upper Mesopotamia during the Roman Period.” In Zeugma, Vol. 6: La Syrie romaine: Permanences et transferts culturels, edited by Catherine Abadie-Reynal and Jean-Baptiste Yon, 123–37. Lyon: Maison de l’Orient et de la Méditerranée Jean Pouilloux, 2015.

Palme, B. “Die classis praetoria Misenensis in den Papyri.” In Festschrift für Luciana Aigner-Foresti zum 70. Geburtstag am 30. Juli 2006, edited by P. Amann, M. Pedrazzi, and H. Taeuber, 294–96. Vienna: Holzhausen, 2006.

Pellat, Ch., “al-Marzūqī.” The Encyclopaedia of Islam, Second Edition, edited by P. Bearman, Th. Bianquis, C.E. Bosworth, E. van Donzel, and W.P. Heinrichs. Leiden: Brill, 2012. Accessed August 29, 2017. http://dx.doi.org/10.1163/1573-3912_islam_SIM_4992.

Podossinov, A. Vostochnaya Evropa v rimskoi kartograficheskoi traditsii, teksty, perevod, kommentarii. Moscow: Indrik, 2002.

Rathmann, M. “The Tabula Peutingeriana and Antique Cartography.” In Brill’s Companion to Ancient Geography. The Inhabited World in Greek and Roman Tradition, edited by S. Bianchetti, M. Cataudella, and H.-J. Gehrke, 335–62. Leiden: Brill, 2015.

Rivolta, P. “Miti letterari e programmi politici nelle orazioni ‘giulianee’ di Libanio.” Clio 21 (1985): 525–46.

Rosivach, V. J. “The Romans’ View of the Persians.” The Classical World 78, no. 1 (1984): 1–8.

Rübekeil, L. Diachrone Studien zur Kontaktzone zwischen Kelten und Germanen. Vienna: Verlag der Österreichischen Akademie der Wissenschaften, 2002.

Savant, S. B. The New Muslims of Post-Conquest Iran: Tradition, Memory, and Conversion. Cambridge Studies in Islamic Civilization. New York: Cambridge University Press, 2013

Schneider, R. M. “Orientalism in Late Antiquity. The Oriental in Imperial and Christian Imagery.” In Ērān ud Anērān: Studien zu den Beziehungen zwischen dem Sasanidenreich und der Mittelmeerwelt. Beiträge des Internationalen Colloquiums in Eutin, 8.–9. Juni 2000 edited by J. Wiesehöfer, and P. Huyse, 241–78. Stuttgart: Steiner, 2006.

Schuol, M. “Indien und die großen Flüsse auf der Tabula Peutingeriana: die östliche Oikumene zwischen paganer und christlicher Kartographie,” Orbis Terrarum 14 (2016): 92–155.

Smith, G. R. “The political history of the Islamic Yemen down to the first Turkish invasion (1–945/622–1538).” In Studies in the Medieval History of the Yemen and South Arabia, edited by G. Rex Smith, 129–39. Variorum Collected Studies Series. Aldershot: Ashgate, 1997.

Smith, G. R. A Traveller in Thirteenth –Century Arabia: Ibn al-Mujāwirs Tārīkh al-Mustabṣir. Works issued by the Hakluyt Society. Edited by W.F. Ryan, Gloria Clifton, and Joyce Lorimer. Third Series, No. 19. London: The Hakluyt Society, 2007.

Smith, R. “The Casting of Julian the Apostate ‘in the Likeness’ of Alexander the Great: a Topos in Antique Historiography and its Modern Echoes.” Histos 5 (2011): 44–106.

Steingass, F. J. A Comprehensive Persian-English Dictionary: Including the Arabic Words and Phrases To Be Met with in Persian Literature Being Johnson and Richardson’s Persian, Arabic, and English Dictionary. London: Routledge, 1988.

Stenger, Jan R. “Libanius and the ‘Game’ of Hellenism.” in Libanius: A Critical Introduction, edited by L. Van Hoof, 268–92. Cambridge: Cambridge University Press, 2014.

Szombathy, Z. “Genealogy in Medieval Muslim Societies.” Studia Islamica 95 (2002): 5–35. Accessed October 17, 2014. http://www.jstor.org/stable/1596139.

Tabacco, R. Itinerarium Alexandri. Testo, apparato critico, introduzione, traduzione e commento. Turin: Leo S. Olschki, 2000.

Talbert, J. A. “Peutinger’s Roman Map: The Physical Landscape Framework.” In Wahrnehmung und Erfassung geographischer Räume in der Antike, edited by M. Rathmann, 221–30. Mainz: Verlag Phillip von Zabern, 2007.

Talbert, J. A. Rome’s World: The Peutinger Map Reconsidered. Cambridge–New York: Cambridge University Press, 2010. Extra information at the publication’s website. Accessed Fenruary 27, 2018. http://www.cambridge.org/us/talbert/.

Teitler, H. C. “Visa vel lecta? Ammianus on Persia and the Persians.” In The late Roman World and its Historian: Interpreting Ammianus Marcellinus, edited by J. W. Drijvers, and D. Hunt, 216–23. London–New York: Routledge, 1999.

Tomaschek, W. “Zur historischen Topographie von Persien, 1: Die Straßenzüge der Tabula Peutingeriana.” In Sitzungsberichte der philosophisch-historischen Classe der kaiserlichen Akademie der Wissenschaften, vol. 102, 145–231. Vienna: Biblio Verlag, 1883.

Weber, E. Tabula Peutingeriana: Codex Vindobonensis 324. Graz: Akademische Druck- u. Verlagsanstalt, 1976.

Weber, E. “Zur Datierung der Tabula Peutingeriana.” In Labor Omnibus Unus. Gerold Walser zum 70. Geburtstag dargebracht von Freunden, Kollegen und Schülern, edited by H. E. Herzig, R. Frei-Stolba, and G. Walser, 113–17. Stuttgart: Steiner, 1989.

Weber, E. “Areae fines Romanorum.” In Altertum und Mittelmeerraum: Die antike Welt diesseits und jenseits der Levante: Festschrift für Peter W. Haider zum 60. Geburtstag, edited by R. Rollinger and B. Truschnegg, 219–27. Stuttgart: Steiner, 2006.

Weber, E. “Die Datierung des antiken Originals der Tabula Peutingeriana.” Orbis Terrarum 14 (2016): 229–59.

Whitehouse, D. “Excavations at Siraf: First Interim Report.” Iran 6 (1968): 1–22.

Whittaker, C. R. “Mental Maps and Frontiers: Seeing like a Roman.” In Thinking Like a Lawyer: Essays on Legal History and General History for John Crook on His Eightieth Birthday, edited by P. McKechnie, 81–112. Leiden: Brill, 2002.

Wiebke, V. Das Imperium Romanum und seine Gegenwelten: die geographisch-ethnographischen Exkurse in den „Res Gestae“ des Ammianus Marcellinus. Berlin–Boston: Walter de Gruyter, 2013.

Wiemer, H.-U. Libanios und Julian. Munich: Beck, 1995.

Winter, E., and Beate Dignas. Rom und das Perserreich: Zwei Weltmächte zwischen Konfrontation und Koexistenz. Berlin: Akademie Verlag 2001.

 

Some Thoughts on the Translation and Interpretation of Terms Describing Turkic Peoples in Medieval Arabic Sources*

Zsuzsanna Zsidai

Hungarian Academy of Sciences, Research Centre for the Humanities

The identification of the various peoples who lived on the medieval Eurasian Steppe has always been an engaging problem among scholars of the early history of this territory. The Arabs came into contact with Central Asian peoples from the beginning in the seventh century, during the course of the Islamic conquest. Hence, one finds many details about the peoples of the Steppe in the Arabic sources.

The Arabic geographer Ibn Rusta mentions the Hungarians among the Turkic peoples in the beginning of the tenth century. However, according to the Arabic sources, there were many Turkic tribes or peoples in different regions, such in Ferghana, Khorasan, Transoxania, Samarqand, and near Armenia. Based on this fact, the term “Turk” can be interpreted in different ways. My aim is to indicate some of the difficulties concerning the translation and interpretation of the terms referring to peoples or tribes, such as “jins” and “qawm,” and to give some examples of occurrences of the ethnonym “Turk” in medieval Arabic texts.

I begin with a discussion of the relevant methodological questions and then argue that the designation “Turk” should be used more cautiously as a group-identifying term in the wider context of the early Medieval world of the Eurasian Steppe.

Keywords: Turks, ethnonyms, Eurasia, Arabic sources

Introduction

The Arabs conquered Central Asia in several waves of attacks and finally overthrew the Chinese forces at the Talas river in 751, so they annexed Transoxania to the Caliphate. First, one must highlight the importance of contacts between various peoples and cultures in Eurasia and the long-durée changes that shaped the history of the region. However, this would go beyond the framework of this paper. We can find traces of the meeting of Arab and Eurasian peoples and cultures in the archaeological heritage but also in the medieval Islamic geographical and historical literature. These sources contain very important information about the early medieval history of these territories and their peoples, but one must keep in mind that, if we seek to arrive at an understanding of the wider context of the region’s history, we need to consider the Turkic, Chinese, Uighur, and Persian sources as well.1 Many of the Steppe peoples who lived in different regions (such in Ferghana, Khorasan, Transoxania, Samarqand, and near Armenia), are referred to as Turks in the Medieval Islamic texts. It is therefore sometimes difficult to identify the various “Turkic” peoples in the sources. My research focuses on early Hungarian history (by which I mean the period before the eleventh century), to which this issue is relevant because the Hungarians were referred to in the sources primarily as Turks, but the “Turk problem” is a very important and fascinating question in the wider context of the world of the early Medieval Eurasian steppe too. In the following, I would like to emphasize that as an Arabist, I will examine these questions on the basis of Arabic sources exclusively. One must begin with the first question: who were the peoples referred to as Turks in the sources, and which parts of the Steppe did they inhabit?

If we speak of Turkic peoples, even if we take into consideration their skills in military affairs and their emergence into the politics of the Islamic caliphate during the centuries following the Arabic conquest, it is interesting to see how the nomadic, barbarian, and pagan Turkic peoples became the defenders of Islam and the Caliphate. Yehoshua Frenkel correctly points out that the image of the Turks has changed over time, and he assumes that descriptions of the Turks in Arabic sources can be divided into two main periods, the early stage contacts (ca. 650–830, when the peoples of the Steppe were characterized as barbarians) and the later period (830–1055), during which their image evolved into that of the noble savage.2 He analyzes the second period in his article using a wide array of sources. His examinations and recent translations3 of texts about the Turkic peoples are very important and highly valuable, giving some insights into their history and showing their main characteristics in the medieval Arabic texts. Nevertheless, many questions remain concerning shifts in the descriptions of the Turks in the Arabic sources. Hopefully, future studies will pay attention to this subject as regards the early Islamic age, too.

The Question of Group-identifying Terms

Before speaking of the problem of the identification of these peoples,4 one must raise questions related to the usefulness and limitations of group-identifying terms in general. This is a very complex problem, which concerns not only the translation of words, but also interpretations which are subjects of the fields of history and anthropology. If one reads about Turkic or any other kinds of peoples in the Medieval Arabic geographical or historical works, one finds many expressions and sentences resembling the following two examples:

 

“at-turk ummatun ʿaẓīmatun kathīratu al-ajnāsi wa al-anwāʿi kathīratu al- qabāʾila wa’l-afkhādhi”

“The Turks are a great people and consist of many kinds and varieties, many tribes and sub-tribes”);5 (Trans. Minorsky)

“wa fīhi ayḍan jinsun min al-ṣaqāliba” 6 (“and [in the Caucasus] [dwells] a kind of Slavic peoples too”). “Wa’l-majghariyya jinsun min al-turk” 7 (“The Hungarians are a kind of Turkic people”).

But the question arises, which social/ethnic groups/tribes or peoples are mentioned among the Turkic peoples by the authors?8 When reading about the early Hungarians or any other kind of Turkic peoples, this can be confusing, even if one keeps in mind that the identification of ethnicity is another general issue.9 In order to further an examination of the categories of “Turkic” peoples, it is essential to consider the interpretation of the word “jins,” and other terms which are used in the medieval Arabic texts to designate peoples or tribes should also be interrogated. I list the most specific terms found in the sources.

I would like to begin by emphasizing that a full discussion of the problem of “tribes” lies beyond the scope of my research and this paper. However, it is important to summarize the main methodological questions, which are strongly connected with the focus of this inquiry, namely the questions relating to translations and interpretations of words and terms designating various social groups in our written sources. The problem of tribe and its translations may be too broad and complicated in part because it demands interdisciplinary work from the fields of philology, anthropology, and history, which would be a complex undertaking.10 It has become almost a commonplace in anthropology that the main problem of the tribe is that it is a “magical word,”11 and it is hard, if not impossible, to define what it means exactly.12 The meaning of “tribe” can be quite different and can shift over time, depending on a wide variety of factors, such as territory, the exact period of time in question, or the origins of the author and whether or not the term is used to denote a particular fluid society.13 This also means that in most cases, it is a difficult to translate and interpret the terms or nouns describing groups, peoples, or tribes, and in some ways, the mapping of these social groups, if they can be mapped at all, is strongly connected with the ethnographers’ (or translators’) fictions.14 Despite the serious methodological issues, it might be worth taking into consideration the anthropologists’ notes and considering how their findings could be used in historical research. Of course, many methodological problems arise, for instance the question of extrapolation of sources,15 such as the case of the word “īlāt.” This word has been applied to the tribal, pastoral, nomadic population, but it is not found in the medieval Persian records.16

Surprisingly, it was social anthropologist David Sneath who raised the problem of the interpretation of these terms some years ago and suggested that “specialists in other fields” should think about the problem of translations.17 While Sneath is not a philologyst, he recognized this fundamental issue concerning the Mongol era and Persian texts, and he found that the word “qawm” in Rashīd al-Dīn’s work does not mean “tribe” or “lineage.”18 With regards to the interpretation of group-identifying terms, I assume that his arguments are persuasive. Christopher P. Atwood also emphasizes that there is no comprehensive study of the terms used to designate various groups in Rashīd al-Dīn’s work. For example, the word “qawm” seems to be regularly applied to any Turco-Mongol social group, so it is not possible to specify its meaning. At the same time, he points at the difficulties of the simultaneous usage that were common with reference to the interpretation of his material as ethnographical research into the Pre-Chinggisid Mongols, which is a very important observation.19

There are articles demonstrating the unambiguousness of the usage of words like “peoples” or “tribes” in various sources. A few decades ago, Richard Tapper mentioned the problem of interdisciplinary studies in this field, and he emphasized that historians and philologists translate and interpret words like “qabīla”,” ṭāʾifa”,” qawm” as tribes many times but without knowing how they were actually used by the authors.20 To my knowledge, there is also no comprehensive study examining the terms mentioned in various sources and originating from different regions, like the Arabian Peninsula, the Middle East, or Central Asia. However, in the field of Oriental studies, recently some articles raised this issue concerning interpretations of tribe in written sources from the perspective of the representation of communities,21 or they examined the terms used to designate nomadic peoples.22 In a recent article, Johann Heiss and Eirik Hovden analyzed and compared the terms describing tribes or social groups in al-ʿAlawī’s (ninth-tenth century) and al-Hamadhānī’s (tenth century) works, and they found that al-­ʿAlawī used mostly the term “ʿashīra” when speaking of tribes or groups of peoples, while interestingly, this word is not found in al-Hamadhānī’s genealogical work. They highlighted that al-ʿAlawī was of north Arabian origin, while al-Hamadhānī belonged to the south Arabian peoples, and this may have been one of the reasons why they used different terms to describe various social groups.23 The general use of group-identifying terms in the case of Turkic or other peoples differs significantly from this, and I would like to add some examples from the Arabic sources describing Turkic peoples and point out some difficulties concerning the translations of group-identifying terms. Of course, it is impossible to understand how Turks identified themselves on the basis of the Arabic sources, as these sources are external and they depict these peoples mostly as nomads, barbarians, or infidels,24 but this could be the subject of another paper.

 

Jins

In the medieval Arabic sources, one finds many terms designating Turkic peoples, such as jins, umma, qawm, qabīla, and ṭāʾifa. The lexicons of Régis Blachère or Edward William Lane or even simply Ibn al-Manẓūr’s dictionary give a good idea of the diversity of meanings of these words. The most common word one finds in these descriptions of nomadic Turkic peoples is jins. This term basically means a kind or class within a higher-order thing, for example in the case of animals and peoples:

 

”al-ḍarbu min kulli shaʾyin, wa huwa min al-nāsi wa min al-ṭayri…”

”[this word means the] kind of everything, such as the [kind of] people or birds…”

 

In this sense, the modern Arabic dictionary later also gives “nation” as one possible meaning. Jins might be a loanword from the Greek γένος and Latin genus (though these terms do not have the same meaning), and it usually refers to a species within a genus.25 “Jins” can also refer to pagan or barbarous peoples, or other ethnic groups.26 If one takes a closer look at the geographical sources, one sees that the term “jins” can designate smaller or larger groups of people, including Turks, Chinese, Indian peoples, or Slavs:

“jinsun min al-turk,” [they are]a kind of Turkic peoples;

“ajnāsi al-turk,” groups of Turkic peoples;

“jinsun min al-saqāliba,” a kind/group of Slavic peoples;

“wa ajnāsun min al-turk badū yusammūna al-w.l.n.d.riyya,”27 and [some] kinds of the nomadic Turks called al-w.l.n.d.riyya

 

This word is translated many times as race28 or tribe, but the word race should not be used anymore, especially in these kinds of translations, and “jins” usually denoted a group or kind of peoples, which is very common in the geographical literature. For example, al-Masʿūdī (†H 345/956) writes the following in his Kitāb al-tanbīh:

 

The fifth group of peoples (ummatun)29 consists of [various] kinds of Turkic peoples (ajnās al-turk), and among them are the kh.r.l.khiyya, the ghuzz and kīmāk, and the ṭughuzghuz and the khazar. [The Khazars] are called sabīr in Turkish and al-khazarān in Persian, and they are a kind of Turkic peoples (jinsun min al-turk) who are settled [people], and their name was Arabized. It is related that the Khazars and other [kinds of Turkic peoples] have one common language, and they have one king.30

As one sees, the term “jins” refers here to a larger group or a kind of Turkic peoples. At the same time, in the work of al-Marwazī, V. Minorsky translated the word “jins” in some places as tribes, but it is possible that the author meant tribes:

 

”wa ʿan al-yasār al-Ṣīn ʿindā maṭlāʿi al-shamsi al-ṣayfi khalqun kathīratun fīmā bayna al-Ṣīn wa al-khirkhīz wa hum ajnāsun lahā asāmin mithla Abrmr (?), Ḥwrnyr (?), Tūlmān (?), F.rāḥnklī (?), Yāthī (?), Ḥynāthī (?), Būbūʿnī (?), B.nkū (?), Fūrī (?).”

”To the left of China towards the summer sunrise, between China and the Kyrgyz, there is a large population. They are tribes with names such as Abrmr (?), Ḥwrnyr (?), Tūlmān (?), F.rāḥnklī (?), Yāthī (?), Ḥynāthī (?), Būbūʿnī (?), B.nkū (?), Fūrī (?).” (Trans. by Minorsky)

V. Minorsky noted that the transcription of these names is conventional and cannot be relied upon, but this is another problem concerning the interpretation of the sources related to the history of these peoples.31

It is also hard to determine what kind of social or relational connection this word had. An interesting example of the use of term “jins” is found in the work of Ibn al-Athīr (†1233), in which he contends that the Tatars wanted to ally with the Kipchaks against the Alans in 1222, and they based their argument on the “fact” that they and the Kipchaks originated from the same “jins”, but the Alans did not. As one later sees, the Tatars used this only as a reason to attack the Kipchaks.32 Here the word “jins” seems to refer to a kind of kinship connection between the Kipchaks and the Tatars, but we know little of this, and in the end, obviously, it meant nothing to the Tatars. Emphasis in this case can be placed rather on the argument itself: how did they make friends out of their enemies, and how did they use this during the negotiations?

 

Umma

Another term which is often found as a designation of different kinds of peoples is the plural of word umma: umamun. The word “umma” refers primarily to the Muslim religious community, but of course, it can have different meanings in various sources. Michael Cooperson examined uses of the term “umma” on the basis of al-Masʿūdī’s work. He suggests that the term was used to denote peoples, nations, or communities as well, and its attributes were in flux. If one is speaking of larger communities, such as nations, one could mention the Persians, the Byzantines, the Chinese peoples, Turks etc. among the major umam of the ancient world in the historical and geographical literature. Cooperson also assumes that al-Masʿūdī was well aware of the difficulties of the reconstruction of each umma’s history.33 Heiss and Hovden concluded that in the singular, “umma” meant mostly the universal Muslim community, and in the plural (umamun) referred to the many peoples from different part of the world and among them to the Muslim community’s pagan and heterodox enemies. They give an example from al-Idrīsī’s (†1165) work, in which the author used this term to designate peoples along the East African coast or the Turkic peoples of Central Asia.34 One finds instances of this in other sources too, for example when we read the following about the Turkic peoples in al- Marwazī’s work:

 

“wa minhum khirghiz wa hum ummatun kathīratun,” and among [the Turks] there are the Kyrgyz people, who are a great people”

“wa ʿalā yamīn hāʾulāi al-kīmākīya thalāthu umamin yaʿabudūna al-nayyirān wa’l miyāha” and on the right side of these Kimeks, there are three kinds of peoples who adore the sun and the moon and the waters”35 (Trans. by Minorsky)

It is worth noting that he uses the words “umam” and “ajnās” (i.e. as plural forms) quite often, but it is not clear what the difference is between these terms exactly. In another passage, al-Masʿūdī mentions the Burtās people as “ummatun ʿaẓīmatun min al-turk,” or “a community or group from the Turkic peoples.”36 As one can see, this word denoted primarily larger or smaller groups of peoples out of the Muslim communities in terms of Turkic peoples.

 

Qawm

The other term designating larger or smaller groups is qawm or aqwām in the plural. This word can be found in an array of geographical works. For example, al-Marwazī mentions the Magyars as “qawmun min al-turk,” or “the Majgharī are a Turkish people” in V. Minorsky’s translation, which is the same in Ibn Rusta’s work, though he refers to them as a “jins”, not as a “qaum”.37 In another passage, he writes about the Pechenegs:

 

“wa’l-bajnākīyya qawmnun sayyāratun,” or “the Pechenegs are wandering people.”38 Ibn Faḍlān also uses this term in his work: “baladu qawmin min al-atrāk yuqālu lahum al-bāshghird,” or “the land of a kind of Turkic peoples called Bashkirs.”39

Qabīla

There are other words which the authors used primarily to refer to smaller groups of peoples, such as tribes. One of these words is qabīla (in the plural qabāʾilu). As Heiss and Hovden highlight, this is not a term denoting exclusively Arabs, though one finds other mentions of non-Arab peoples, mainly in travelers’ accounts, in which they write about non-Arab-or Islamic lands.40 This is the case with the Turkic peoples too. For example, Ibn Faḍlān refers to the Oghuz peoples as “tribes”: “qabīlatun min al-atrāk yuʿrifūna bi’l-ghuzziya,” or “a tribe of Turkic peoples known as Oghuz.”41 He also mentions the tribes of infidel peoples:

 

“fa-baynakum wa bayna hadhā al-baladi alladhī tadhkurūna alfu qabīlatin min al-kuffār, or “and between you and the land, which you have mentioned, there are one thousand tribes of infidels.”42

But this word can be found in many other works too, including for instance Ibn al-Faqīh al-Hamadhānī’s description:

 

“wa Yājūj wa Mājūj arbʿa wa ʿishrūna qabīlatan fa-kānat qabīlatun minhum al-ghuzzw wa hum al-turk,” or “And Gog and Magog had 24 tribes (?) and there was a tribe (?) among them, the Oghuz, and they are the Turks.”43

Ṭāʾifa

Another term which was widely used to designate tribes in the Arabic sources is ṭāʾifa (pl. ṭawāʾifu). This basically means a part of something (“juzʾun min al-shayʾi”) and also a group of people (”jamāʿatun min al-nāsi”) numbering less than one thousand,44 and in this sense, as since it designates a smaller group of peoples, the word can be translated as tribe. This word describes many groups of peoples or tribes in the Middle East and Central Asia, and it has been studiously analyzed in the anthropological scholarship. For example, the term “qawm” and “ṭāʾifa” are widely used today in Iran and Afghanistan and they can refer to various levels of the social organization of a group of peoples like tribes, groups, and the like.45

This word can denote Turkic peoples in the Arabic geographical literature, for example in al-Bīrūnī’s or Ibn Faḍlān’s works:

 

“wa ḥawlahu (al-baḥr al-khazar) ṭawāʾifu min al-turk wa-al-rūs wa-al-ṣaqlab,” or “and around the (Caspian Sea) dwell groups of the Turkic, Rūs and Slavic peoples”; 46

“wa raʾynā ṭāʾifatan minhum taʿbudu al-ḥayyāta wa ṭāʾifatan taʿbudu al-samaka wa ṭāʾifatan taʿbudu al-karākīya,” or “and we saw a group of them, which worshiped the snake, a group, which worshiped the fish, and [another ] group, which worshiped the cranes.”47

Thus, this term can refer to tribes or different kinds of peoples in the sense of the Arabic word nawʿ at the same time.

Other Terms and the Problems of Interpretations

There are other words like “jīl” which can also stand for smaller or larger groups of people or tribes, but it is only rarely used in descriptions of the Turkic peoples. Blachère suggested it refers primarily to larger groups of peoples, like the Chinese, the Turks, etc. as is mentioned in Ibn Manẓūr’s dictionary,48 but Lane found that “jīl” can also refer to tribes, and in al-KÁshgharī’s DīwÁn one finds the same assertion, although no Turkic word is given as an equivalent of this term.49 It is worth noting that the term ʿashīra (pl. ʿashāʾir), which can denote smaller sub-tribes of qabīla,50 is rarely used in the sources to denote Turkic peoples, and indeed I myself have not seen it used once to denote Turkic peoples.

On the basis of the examples mentioned above, one can conclude that the translation of these words can be very difficult and uncertain, which means that ultimately the translation is an interpretation of the terms. One comes across several examples of this when reading about the history of the Eurasian Steppe, because in the sources there are various words which are consistently

translated as tribe. For instance, there is a fascinating article about the people of Nūkarda, and there are some places where the translations of “tribes” are confusing. The author, Turkologist Peter Golden, translates both “jins”51 and “jīl” as tribe.52 However, if one takes a closer look at the given text, one sees that these words could denote larger groups of peoples, or at least they could refer to different kinds of peoples that were described by al-Masʿūdī there. “al-jīlu al-awwalu minhum yuqālu lahum bajnā, thummā yalīhi ummatun thāniyatun yuqālu lahā bajghird, thumma yalīhā ummatun yuqālu lahā bajnāk … talīhā ummatun ukhrā yuqālu lahā Nūkarda.” He translates this as follows: “The first tribe is called bajnā. Near to them is the second people, who are called bajghird, and near them is a people, the bajnāk, … near them is the last of these peoples, the Nūkarda .”53 I would venture the contention that it is not immediately obvious that, when using the word ”jīl”, he meant tribe, as the text is a listing of the peoples living in the Caucasus. The other thing is that ummatun ukhrā does not mean the last of these peoples, but rather can be translated as follows: “[they are followed by] another [group of] peoples called Nūkarda.”54 One notices the same thing if one also reads the translation of anwāʿ (kinds, sorts, species) as tribes,55 though they do not have this meaning.56 Here the author quotes al-Masʿūdī’s historical work, in which he mentioned the Black Sea: “al-burghār wa al- rūs wa bajnāk, bajghird wa hum thalāthatu anwāʿin min al-turk.”57 He translates as follows: “The Burghar, the Rus, the *Pacänǟ, the Päčǟnak and the Bajğird, (the latter) are three tribes of the Turks.”58 The word nawʿ cannot mean tribe here, so they are three kinds of Turkic peoples. Moreover, al-Masʿūdī wrote about the Baḥr Nītas in the first instance, describing them as the sea of the people of Burghar, the Rus, the *Pajänä, the Päjänak, and the Bajğird. Golden, however, assumes that he is speaking of three tribal organizations.59 I would suggest that the sentence should be translated as follows: “And as the astrologers from among the holder of astrological tables and other [astrologers] among the elders say, the sea of al-Bulghar and al-Rūs [and B.j.nÁ and B.j.n.Ák and Bgh.r.d—and they are three kinds of the Turkic peoples] is the Sea of Nīṭ.sh. (the Black Sea)”60 Adding to this, al-Masʿūdī mentions the Burghar as a kind of Slavic people using the term “nawʿ” (nawʿ min al-Ṣaqāliba) in his geographical description, which does not denote tribes there.61 Finally, in the same article there is a sentence in which one finds the word “jins”, but it has not been translated at all.62 The article is still highly valuable, but the translator thus can confuse the reader, even if he also correctly noted later, in another passage, that he is uncertain as to how to translate the word “jins”.63 In the recent translations of excerpts about Turkic peoples in the Arabic sources, Frenkel found the translation of these group-identifying words as hard as in the case of Ibn al-Faqīh al-Hamadhānī’s work.64

In conclusion, how these terms are translated is important. If one examines the history of the Steppe peoples, it does matter whether they are referred to as peoples or tribes, especially if one seeks to analyze their state/tribal organization.65 Unfortunately, in most cases one does not find descriptions of these terms that are as detailed and clear as the ones found in the Arabic-Turkish dictionary of al-Kashghārī, in which he describes them quite precisely.66 On the basis of the abovementioned examples, and because these words denote tribes or particular fluid social groups, I would like to argue that we should use “jins”, “qabīla”, “qawm”, or “ṭāʾifa” etc. as group-identifying terms more cautiously in the wider context of the early medieval world of the Eurasian Steppe. Moreover, one also has to consider that it is not possible to apply “modern” (or Western) terms like nation for the description of the communities of the medieval (and eastern) Steppe. With regards to the Arabic sources, Heiss and Hovden have recommended further comparisons and analyses of various texts from different regions in a historical context which would be based on source criticism.67 I can only highlight the importance of their suggestion as it concerns the sources on the Turkic peoples of the Eurasian steppe.

The Ethnonym Turk and Problems with Its Use

In the following, I raise the problem of the interpretation of the ethnonym Turk. Narratives of early Hungarian history (i.e. the period before the eleventh century) offer many examples of the problems with the use of this term because of the scarcity of sources and also because the early Hungarians were nomadic, so they were mentioned as Turks not only in the Islamic sources but very often in Latin and Greek sources too. Studies on the so-called Turkic peoples are popular, but there are few works and little research on the history of the Turks which rely on the Arabic sources before and by the time of early Islam because this period of the Turkic people’s history is poorly documented. The problem has been discussed in the international research,68 however, and it is clearly important to consider carefully how the sources use the term “Turk” (i.e. to which groups of people or peoples do they apply it).69 If one only takes the English translation of al-Ṭabarī’s chronicle (Taʾrīkh al-rusul wa’l-mulūk) into account, one will have difficulties regarding the identification of “Turks.”70 Apart from the chronicle of al-Ṭabarī, one could also mention the case of the Khalaj Turks. According to an article by Miklós Maróth, who examined this question on the basis of the al-Balkhī tradition, the Khalaj Turks lived between the steppe of al-Dāwar and Ghazna. Maróth agrees with al-Khwārizmī’s conclusion that they were the descendants of the Hephthalites if it was true that the Hephthalites were Turkic.71 But this assumption is related to the problem of the Hephthalites (hayāṭila in the Arabic sources), which is another interesting subject of debate among scholars at the moment.72

With respect to early Hungarian history, which is strongly connected to the history of the Eurasian Steppe, unfortunately in some cases it is far from clear that a given source which mentions “Turks” has any connection to the history of the early Magyars, and this raises the problem of “Turk” as an ethnonym, too.73 One example is found in an interesting passage in Ibn Rusta’s work, which derives from Hārūn ibn Yaḥyā, who lived in Constantinople and described the Byzantine Empire and its neighbors. The passage in question goes back to the second half or the end of the ninth century. Hārūn ibn Yaḥyā mentions Turks as guards of the emperor.74 On the basis of an analysis of the De administrando im­perio (DAI), which was edited by Emperor Constantine VII of Byzantium (913–59) in the middle of the tenth century, and the work of Ibn Rusta, Joseph Marquart concluded that these Turks were Turks from Ferghana (Φαργάνοι). However, he quoted another passage from the DAI in which the term “Turks” (Τοῦρκοι) refers to the Turks of Ferghana, the Khazars, and other soldiers who might have been Hungarians.75 Some historians have concluded that this fragment refers clearly to the early Hungarians, but I do not see any clear evidence in support of this conclusion.76

Another example of the misinterpretation of ethnonyms is the case of Samanid IsmÁÝÐl ibn Aḥmad’s raids against Taraz in 893. Al-TabarÐ’s account of this event is mentioned in the historical sources on the Hungarian conquest because some of the Hungarian historians and archaeologists thought it was this raid which caused the Pechenegs’ raid against the Hungarians, which may have prompted the Hungarians to migrate into the Carpathian Basin in 896.77 If one takes a closer look at the sources, however, one sees that al-Mascūdī’s work, in which he wrote more about the raids and fights on these territories, unfortunately has been lost, and no sources have been found describing this raid as the starting point of an eastern-western migration of the Karlukhs towards to the Oghuz people who dwelled near the Aral Sea. Instead, according to the sources, part of the Karlukh people moved to Kasghar, and this city lies not to the west, but to the southeast of Taraz. Moreover, if we look at the map of this raid as it is reconstructed in the secondary literature, there is no clear explanation why Bukhara would have been the starting point of the raid. Al-NarshakhÐ writes that the Samanid emir returned to Bukhara with the captives and booty, but there is no mention in any of the sources of the specific site from where the raid was launched, so in all likelihood, this argument was based only on the fact that IsmÁÝÐl ibn Aḥmad was the emir of Bukhara by that time.78 Unfortunately, the abovementioned problems of translation and interpretation notwithstanding, one of the most important migration hypothesis concerning the early Hungarians is based on this argument. After having studied the related sources, however, I have come to the conclusion that we cannot consider this raid the starting point of a greater migration, at least not in the case of the Hungarian conquest. Rather, it was in all probability an important event in a longer border fight between the pagan Turkic/Nomadic peoples and the Caliphate. These fights are important from the perspective of the history of the steppe, and I find Deborah G. Tor’s argument interesting. Tor contends that there are not many notes on these raids against the Turks because these fights resulted in great losses and deficits for the Caliphate.79 Whatever the truth is, it would be worthwhile to reevaluate our sources with regards to the Arabic conquest of Central Asia as well. Apart from the problems of the sources on the early history of the Hungarians, the use of ethnonyms is confusing in other texts too. Sometimes, a name does not refer to a people but rather to the territory where they live, for example al-Iṣṭakhrī mentioned the name Burṭās (who were described by al-Mascūdī as Turkic peoples as well)80 as an “umma” and later as “nÁhiye”: “Burṭās is the name of a region (nāḥiya), like the Rūs, and the Khazar and the Sarīr, which are all the names of countries (mamlaka) and not the names of cities or peoples.”81 However, the term Khazar was used to denote peoples in the beginning of the same work: “as regards Khazar, it is the name of this kind of peoples (innahu ismun li-hadhihi al-jinsi min al-nÁsi).”82 Claus Schönig has also noted the ambiguousness of term Turk in al-KÁshgharÐ’s DÐwÁn, and he concludes that the term Turk denoted 1) the Turkic people as a whole, 2) the non-Oghuz peoples (in remarks on the Oghuz dialect), and 3) a part of the core population of the Karakhanid state, i.e. the Čigil.83

Further examples of the use of the term “Turks” could be listed, but they would not add to the core argument of my inquiry. Another important factor is the question of the “Turkicization” and Islamization of the territory where “Turks” had lived earlier. A decade ago, Sören Stark published a book which examined this question from the perspectives of archaeology and history,84 and in a later article he noted a problem concerning the early Turkic archaeological material and the interaction between the inhabitants of early medieval pre-Muslim Transoxania. He also noted that, “[t]he actual status of these earliest influences [viz. the middle of the first millennium A.D.] from the Turkish steppes in Transoxania is still poorly understood and consequently a matter of considerable dispute between archaeologists, historians and linguists.”85 In conclusion, each use of the term “Turk” must be interpreted in a wider historical and geographical context, and it is obviously not easy to define which kinds of Turkic peoples were described in the chronicles or the geographical descriptions. Hence, as noted above, historians must be careful with the translations of these ethnonyms.86 The term “Turk” can refer to various kinds of peoples and also tribes, subtribes, or clans, which are mainly nomadic in the Arabic sources. Historians must also keep in mind that the term does not have anything to do with the ethnicity in itself, especially if we speak of the early medieval history of the Steppe.

In my assessment, further study of the uses of ethnonyms like “Turk” is necessary, as is further study of the migration of early nomadic peoples in the historical context of the Eurasian Steppe. This question is interesting not only from the perspective of early Hungarian history, but also as regards the early medieval history of the Steppe. There is still room left for Orientalists, Antropologists and Historians in this field of these studies.

Bibliography

Primary Sources

 

Ibn al-Athīr, cIzz ad-Dīn. Al-Kāmil fi’l-ta’rīkh. Vol. XI. Edited by cUmar cAbd al-SalÁm Tadmurī. Bayrūt: DÁr al-KitÁb al-cArabī, 1999.

al- Bīrūnī, Abū al-Raiḥān Muḥammad Ibn-Aḥmad. ḲÁnūn al- Masʿūdī. Edited by Z. V. Togan. New Delhi: Government of India Publ. 1937.

Ibn al-Faqīh al-Hamadhānī, Aḥmad ibn Muḥammad al-Hamadhānī. Kitāb al-buldān. Bibliotheca Geographorum. Arabicorum V. Edited by M. J. de Goeje. Lugduni Batavorum: Brill, 1967.

V. Minorsky. Sharaf al-zamān: Ṭāhir Marvazī on China, the Turks and India. London: Royal Asiatic Society, 1942.

al-Iṣṭakhrī. Kitāb Masālik wa’l-mamālīk. Bibliotheca Geographorum Arabicorum. I. Edited by M. J. de Goeje. Lugduni Batavorum: Brill, 1927.

V. Z. Togan Ibn Faḍlān’s Reisebericht: Leipzig: Brockhaus, 1939.

Maḥmūd al-Kāšγarī Compendium of the Turkic Dialects. (Dīwān Luγāt at-Turk). Edited translated by Robert Dankoff in collaboration with James Kelly. Part 1. Cambridge, Mass.: Harvard University, 1982.

Ibn Manẓūr, Muḥammad Ibn-Mukarram, L. Lisān al-ʿArab. Vols. I-XVIII. Edited by Muḥammad ʿAbd al-Wahhāb, Muḥammad al-Ṣādiq al-ʿUbaydī. Bayrūt: Iḥyā’ al-turāth al-carabī – mu’assasa al-ta’rīkh al-Ýarabī.

al-Masʿūdī, Abu’l-Ḥasan ʿAlī b. al-Ḥusayn, K. Kitāb al-tanbīh wa’l-ishrāf. Lugduni Batavorum: Brill, 1967.

al-Masʿūdī, Abu’l-Ḥasan ʿAlī b. al-Ḥusayn. Les Prairies d’Or. Edited and translated by C. Barbier de Meynard–Pavet de Courteille. Vol. 8. Paris: Paris Imprimerie Impériale, 1896.

Mirkhond, Mohammed ibn Khwand-shah ibn Mahmoud. Histoire des Samanides. Edited by Charles F. Defrémery. Amsterdam: Oriental Press, 1974.

Ibn Miskawayh, Alī Aḥmad b. Muḥammad b. Yaʿqūb. Tajārib al-’umam. Vol. 4. Edited by KasrawÐ al-Ḥasan. Bayrūt: Dār al-kutub al-ÝilmÐya, 2003.

al-Narshakhī, Muḥammad. Taʾrīkh-i Bukhārā. Edited by Ch. Schefer. Paris: E. Lereoux, 1892.

al-Ṭabarī, Abū Jaʿfar Muḥammad b. Jarīr. Mukhtaṣar taʾrīkh al-rusul wa ’l-mulūk wa ’-l-khulafāʾ. Vol. 4 of 13. Rawāʾia al-turāth al-ʿarabī 3. Edited by De Goeje, M. J. Bayrūt: Khayats, 1960 k.

Ibn Rusta, Aḥmad Ibn-ʿUmar. Kitāb al-aʿlāq al-nafīsa. Edited by M. J. De Goeje. N.p.: Lugduni Batavorum: Brill, 1967.

al-Ṭabarī, Abū Jaʿfar Muḥammad b. Jarīr. Taʾrīkh al-ʾumam wa’l-mulūk. 13 vols. Bayrūt: Dār al-Fikr, 1987.

The history of al-Ṭabarī: an annotated translation. Vol. 15. The crisis of the early caliphate: the reign of ʿUthmān; A.D. 644–656 / A.H. 24–35. Translated by R. S. Humphreys and edited by ʿAbbās Iḥsān. New York: State University of New York Press, 1990.

The history of al-Ṭabarī: an annotated translation. Vol. 18. Between civil wars: the caliphate of Muʿāwiyah; A.D. 661–680 / A.H. 40–60. Translated by M. G. Morony and edited by ʿAbbās Iḥsān. New York: State University of New York Press, 1987.

The history of al-Ṭabarī: an annotated translation. Vol. 22. The Marwānid restoration: the caliphate of ʿAbd al-Malik; A.D. 693–701 / A.H. 74–81. Translated by E. K. Rowson and edited by ʿAbbās Iḥsān. New York: State University of New York Press, 1989.

 

Secondary Sources

 

Atwood, Christopher P. “Mongols, Arabs, Kurds, and Franks: Rashīd al-Dīn’s Comparative Ethnography of Tribal Society.” In Rashīd al-Dīn as an Agent and Mediator of Cultural Exchanges in Ilkhanid Iran, edited by Anna Akasoy, Ronit Yoeli-Tlalim, and Charles Burnett, 223–50. London: Wartburg Institute, 2013.

Bang, W., and J. Marquart. Osttürkische Dialektstudien. Berlin: Weidmann, 1914.

The Chronicle of Ibn al-Athir. 3 vols. The Chronicle of Ibn al-Athir for the Crusading Period from al-Kamil fi’l-Ta’rikh. Part 1–3. English translation and introduction by D. S. Richards. Farnham: Ashgate, 2006–2008.

Bivar, A. D. H. “Hayāṭila” In The Encyclopaedia of Islam, vol. 3, edited by, B. Lewis, V. L. Ménage, Ch. Pellat, and J. Schacht, 303–04. Leiden: Brill, 1971.

Bosworth, C. E. “The Turks in the Islamic Lands up to the Mid-11th Century.” In The Turks in the Early Islamic World, edited by idem, 193–212. Aldershot: Ashgate Variorum, 2007.

Dictionnaire Arabe-Français-Anglais (Langue classique et moderne). Arabic/French/English Dictionary. Edited by Régis Blachère, Charles Pellat, Moustafa Chouémi, and Claude Denizeau. Vol. 3. Paris: Maisonneuve et Larose, 1976.

Cooperson, Michael. “‘Arabs’ and ‘Iranians’: The Uses of Ethnicity in the Early Abbasid Period.” In Islamic Cultures, Islamic Contexts: Essays in Honor of Professor Patricia Crone, edited by Asad Q. Ahmed, Behnam Sadeghi, Robert G. Hoyland, and Adam Silverstein. Leiden: Brill, 2014.

Czeglédy, Károly. “A IX. századi magyar történelem főbb kérdései” [The main questions of Hungarian history of the ninth century]. Magyar Nyelv 41 (1945): 33–45.

Czeglédy, Károly. “A török népek és nyelvek tagozódásának kérdéséhez” [On the question of the grouping of turkish peoples and Turkic languages]. Czeglédy, Károly, Magyar Őstörténeti Tanulmányok, 306–11. Budapest: Magyar Tudományos Akadémia Könyvtára, 1985.

Dankoff, Robert. “KÁšγarÐ on the Tribal and Kinship Organization of the Turks.” Archivum Ottomanicum 4 (1972): 23–44.

Enoki, K. Studia Asiatica: The Collected Papers in Western Languages of the Late Dr. Kazuo Enoki. Tokyo: Toyo Bunko, 1999.

Frenkel, Yehoshua. “The Turks of the Eurasian Steppes in Medieval Arabic Writing.” In Mongols, Turks and others, edited by Amitai, R., and Biran, M., 201‒44. Leiden­‒Boston: Brill, 2005.

Frenkel, Yehoshua. The Turkic Peoples in Medieval Arabic Writings. Abingdon, Oxon–New York: Routledge, 2015.

Frye, R. N., and A. M. Sayili. “Turks in the Middle East before the Saljuqs.” Journal of the American Oriental Society 63, no. 3 (1943): 194–207.

Gibb, H. A. R. The Arab conquests in Central Asia. New York: n.p., 1923.

Golden, Peter. “The people Nūkarda.” Archivum Eurasiae Medii Aevi 1 (1975): 21–35.

Golden, Peter. “The Qipčaqs of Medieval Eurasia: An Example of Stateless Adaptation in the Steppes.” In Rulers from the Steppe: State Formation on the Eurasian Periphery, edited by G. Seaman and D. Marks, 132–57. Los Angeles: Ethnographic Press, 1991.

Golden, Peter. “Review of the Headless State: Aristocratic Orders, Kinship Society, and Misrepresentations of Nomadic Inner Asia, by David Sneath.” Journal of Asian Studies 68, no. 1 (2009): 293–96.

Golden, Peter. “The Turkic World in Maḥmûd al-Kâshgharî.” In The Complexity of Interaction along the Eurasian Steppe Zone in the first Millennium CE, edited by J. Bemmann, and M., Schmauder, 503–55. Bonn: Vor- und Frühgeschichtliche Archäologie, Rheinische Friedrich-Wilhelms-Universität, 2015.

Göckenjan, Hans, and István Zimonyi. Orientalische Berichte über die Völker Osteuropas und Zentralasiens im Mittelalter: Die Gayhani-Tradition (Ibn Rusta, Gardizi, Hudud al-’Alam, al-Bakri und al-Marwazi). Wiesbaden: Harrassowitz, 2001.

Harmatta, János, and Miklós Maróth. “Zur Geschichte der arabisch-türkischen Beziehungen am Anfang des VIII. Jahrhunderts.” In From Alexander the Great to Kül Tegin: Studies in Bactrian, Pahlavi, Sanskrit, Arabic, Aramaic, Armenian, Chinese, Türk, Greek and Latin Sources for the History of Pre-Islamic Central Asia, edited by János Harmatta, 139–44. Budapest: Akadémiai Kiadó, 1996.

Heiss, Johann, and Eirik, Hovden. “The Political Usage of Religious and Non-Religious Terms for Community in Medieval South Arabia: A Comparative Response to Gerda Heydemann’s Chapter.” In Meanings of Community across Eurasia, edited by E. Hovden, C., Lutter, and W. Pohl 61–77. Leiden: Brill, 2016.

Ismail, O. S. A.: “Mu’tasim and the Turks.” Bulletin of the School of Oriental and African Studies 29, no.1 (1966): 112–24.

Kmoskó, Mihály. Mohamedán írók a steppe népeiről: Földrajzi irodalom I/1–3, edited by István Zimonyi. Budapest: Balassi Kiadó, 1997–2007.

Kovács, Kitti. “A Maeotis ingoványai: Történeti-földrajzi vizsgálat antik és középkori szerzők, valamint korai térképek alapján.” Belvedere Meridionale 26, no. 3 (2014): 7–21.

Kradin, Nikolaj N. “The Headless State: Aristocratic Orders, Kinship Society, and Misrepresentations of Nomadic Inner Asia. David Sneath. New York: Columbia University Press, 2007. Review.” Asian Perspectives 21, no. 1 (2012): 130–32.

Kraus, Wolfgang. Islamische Stammesgesellschaften: Tribale Identitän im Vorderen Orient in sozialanthropologischer Perspektive. Vienna–Cologne–Weimar: Böhlau Verlag, 2004.

Kristó Gyula, ed. A honfoglalás korának írott forrásai [The written sources of the Hungarian conquest period]. Szeged: Szegedi Középkorász Műhely, 1995.

Lane, Edward William. An Arabic-English Lexicon Derived from the Best and the Most Copious Eastern sources. Vol. 1. London: Williams & Norgate, 1863.

Lecerf, J.: “ʿAshīra.” In: The Encyclopaedia of Islam. Vol. 1, edited by Lewis, B., Pellat, Ch., and Schacht, J. Leiden: Brill, 1979.

Leder, Stefan: “Nomaden und nomadische Lebensformen in arabischer Begrifflichkeit – Eine Annäherung.” In Nomadismus aus der Perspektive der Begrifflichkeit. Beiträge der 1. Tagung am 11. 7. 2001, 1, edited by S. Leder, and B. Sterck, 1–40. Halle: n.p., 2002.

Lewiczki, Tadeusz. “The Oldest Mentions of the Turks in Arabic Literature.” In The Turks in the Early Islamic World, edited by C.E. Bosworth, 117–19. Aldershot: Ashgate Variorum, 2007.

Maróth, Miklós. “Die Xalağ in den arabischen Quellen.” Acta Antiqua Hungariae 28 (1980): 269–72.

Marquart, J.: Osteuropäische und ostasiatische Streifzüge. Ethnologische und historisch-topographische Studien zur Geschichte des 9. und 10. Jahrhunderts (ca. 840–940). Leipzig: Dieterich, 1903.

Montgomery, J. E. “Ibn Faḍlān and the Rūsiyyah.” Journal of Arabic and Islamic Studies 3 (2010): 1­–25.

Morony, G. M. “Religious Communities in the Early Islamic World.” In Visions of Community int he Post-Roman World, edited by W. Pohl, C. Gantner, and R. Payne, 155–63. Farnham: Ashgate, 2012.

Ormos, István. “A magyar őstörténet arab forrásainak újabb irodalma: Kmoskó Mihály, Hansgerd Göckenjan és Zimonyi István művei” [The latest literature on the Arab sources of Hungarian prehistory: The works of Mihály Kmoskó, Hansgerd Göckenjan, and István Zimonyi]. Hadtörténelmi Közlemények 118 (2006): 733–81.

Ormos, István. “Kiegészítések ‘A magyar őstörténet arab forrásainak újabb irodalma: Kmoskó Mihály, Hansgerd Göckenjan és Zimonyi István művei’ című írásomhoz” [Supplements to my writing entitled “The latest literature on the Arab sources of Hungarian prehistory: The works of Mihály Kmoskó, Hansgerd Göckenjan, and István Zimonyi”]. Hadtörténelmi Közlemények 122 (2009): 1129–57.

Ormos, István. “Remarks on the Islamic sources on the Hungarians in the ninth and tenth centuries.” Zeitschrift für Geschichte der Arabisch–Islamischen Wissenschaften 19, no. 10 (2010): 377–95.

Orywall, E. Die Ethnischen Gruppen Afghanistan: Fallstudien zu Gruppenidentität und Intergruppenbeziehungen. Wiesbaden: Reichert, 1986.

Paul, Jürgen. “Terms for Nomads in Medieval Persian Historiography.” Asiatische Studien: Zeitschrift der Schweizerischen Asiengesellschaft 60, no. 2. (2006): 437–57.

Pohl, Walter. “Conceptions of Ethnicity in Early Medieval Studies.” In Debating the Middle Ages: Issues and Readings, edited by Lester K. Little, and Barbara H. Rosenwien, 13–24. Oxford: Blackwall Publishers, 1998.

Pohl, Walter, and Helmut Reimitz, eds. Strategies of Distinction: The Construction of Ethnic Communities, 300-800. Leiden: Brill, 1998.

Pritsak, Omeljan. “From the Säbirs to the Hungarians.” In Hungaro-Turcica: Studies in honour of Julius Németh, edited by Gyula Káldy-Nagy, 17–30. Budapest: Eötvös Loránd University, 1976.

Schönig, Claus. “On some unclear, doubtful and contradictory passages in Maḥmūd al-KÁšγarī’s ‘DÐwÁn LuγÁt at-Turk.’” Türk Dilleri Araştirmalari 14 (2004): 35–56.

Sinor, Dénes. “The Outlines of Hungarian Prehistory.” Cahiers d’histoire mondiale 4, no. 3 (1958): 513–40.

Sinor, Dénes. “Reflections on the History and Historiography of the Nomad Empires of Central Eurasia.” Acta Orientalia Academiae Scientiarum Hungaricae 58, no. 1 (2005): 3–14.

Sinor, Dénes. “The Establishment and Dissolution of the Türk Empire.” In The Cambridge History of Early Inner Asia, edited by Dénes Sinor, 285–315. Cambridge: Cambridge University Press, 2008.

Sneath, David. “Imperial Statecraft: Arts of power on the Steppe.” In Institutions, Idioms and Technologies of Power in Inner Asia: Introduction to Stets of Mind: Power, Place And The Subject in Inner Asia, edited by David Sneath, 1–13. Bellingham: Western Washington University, 2006.

Sneath, David. The Headless State: Aristocratic Orders, Kinship Society & Misrepresentations of Nomadic Inner Asia. New York: Columbia University Press, 2007.

Sneath, David. “REJOINDERS. A Response by David Sneath to Peter Golden’s Review of The Headless State: Aristocratic Orders, Kinship Society, and Misrepresentations of Nomadic Inner Asia.” The Journal of Asian Studies 69, no. 2. (2010): 658–60.

Sneath, David. “Ayimag, Uymaq and Baylik: Re-examining Notions of the Nomadic Tribe and State.” In Nomad Aristocrats in a World of Empires, edited by Jürgen Paul, 161–85. Wiesbaden: Reichert, 2013.

Somfai Kara, Dávid. “Modern török nyelvű etnoszok kialakulása Közép-és Belső-Ázsiában” [The formation of modern Turkish speaking peoples’ ethnos in Central-and Inner Asia]. In Dentumoger I. Tanulmányok a korai Magyar történelemről [Dentumoger I. Studies about the early Hungarian History], edited by Balázs Sudár, 343–56. Budapest: MTA Bölcsészettudományi Kutatóközpont, 2017.

Southall, A. “Tribes.” In Encylopedia of Cultural Anthropology, vol. 4, edited by David Levinson, and M. Ember, 1329–36. New York: Holt, 1996.

Stark, Sören. “Mercenaries and City Rulers: Early Turks in Pre-Muslim Mawarannahr.” In Social Orders and Social Landscapes, edited by L. M.­ Popova,W. Hartley, and A.T. Smith, 307–34. Newcastle: Cambridge Scholars Publ., 2007.

Stark, Sören. Die Alttürkenzeit in Mittel-und Zentralasien: Archäologische und historische Studien. Wiesbaden: Reichert, 2008.

Szabados, György. “A magyarok bejövetelének hadtörténeti szempontú újraértékelése” [A new assessment of the arrival of the Hungarians from the perspective of military history]. Hadtörténelmi Közlemények 123 (2010): 215–35.

Szabados, György. “Állam és ethnosz a IX–X. századi magyar történelemben” [State and ethnicity in ninth-century and tenth-century Hungarian history]. Acta Historica 135 (2013): 3–24.

Tapper, Richard. “Anthropologists, Historians and Tribespeoples on Tribe and State Formation in the Middle East.” In Tribes and State Formation in the Middle East, edited by P.S. Khoury, and J. Kostiner, 48–73. London: Tauris, 1991.

Tapper, Richard. Frontier Nomads of Iran: A political History of the Shahsevan. Cambridge: Cambridge Univ. Press, 1997.

Tor, Deborah G. “The Islamization of Central Asia in the Sāmānid Era and the Reshaping of the Muslim World.” Bulletin of the School of Oriental and African Studies 72, no. 9 (2009): 279–99.

Van den Bergh, S. “Jins.” In The Encyclopaedia of Islam, edited by B. Lewis, Ch. Pellat and J. Schacht, vol. 2, 550. Leiden: Brill, 1991.

Vaissière, Étienne. “Is There a ‘Nationality’ of the Hephthalites?” Bulletin of the Asia Institute, vol. 17, edited by C.A. Bromberg, 119–31. Bloomfield Hills, 2007.

Vásáry, István. A régi Belső-Ázsia története [The history of ancient Inner Asia]. Szeged: Szeged József Attila Tudományegyetem, 2013.

Vásáry, István. “Hungarians and Mongols as ‘Turks’:” On the Applicability of Ethnic Names.” In Between Byzantium and the Steppe: Archaeological and Historical Studies in Honors of Csanád Bálint on the Occasion of His 70th Birthday, edited by Á. Bollók, G. Csiky, and T. Vida, 537–43. Budapest: MTA BTK Régészeti Intézet, 2016.

Vasiliev, Alexander A. “Harun-ibn-Yahya and His Description of Constantinople.” Seminarium Kondakovianum 5 (1932): 149–63.

Zimonyi, István.“A besenyők nyugatra vándorlásának okai” [The causes of the westward migration of the Pechenegs]. Acta Universitatis Szegediensis. Acta Historica 106 (1998): 129–44.

Zsidai, Zsuzsanna. “Turkok az arab forrásokban ­– problémafelvetés egy népnév értelmezése kapcsán” [Turks in Arab sources: Problems concerning the interpretation of an ethnonym]. Keletkutatás 10, no. 2. (2013): 5–17.

Zsidai, Zsuzsanna, and Péter Langó. “Kunok és alánok Ibn al-Athír munkájában” [Cumans and Alans in the work of Ibn al-Athír]. In Carmen Miserabile: Ünnepi kötet Pálócz-Horváth András 70. születésnapjára [Carmen Miserabile: A commemorative volume on the 70th birthday of András Pálóczy-Horváth], edited by Szabolcs Rosta and György V. Székely, 421–34. Kecskemét: Kecskeméti Katona József Múzeum, 2014.

Zsidai, Zsuzsanna. “Ismāʿīl ibn Aḥmad 893. évi hadjárata és a magyar honfoglalás” [The 893 military campaign of Ismāʿīl ibn Aḥmad and the Hungarian conquest]. In Dentumoger I. Tanulmányok a korai Magyar történelemről [Dentumoger I. Studies about the Early Hungarian History], edited by Balázs Sudár, 109–24. Budapest: MTA Bölcsészettudományi Kutatóközpont, 2017.

* This article was written with the support of the MTA BTK MŐT 28.317/2012 project. I’m grateful for the advice of my teachers and colleagues, and I’d like to thank Miklós Maróth, Walter Pohl, László Tüske, Stephan Procházka, György Szabados, Gabriele Rasuly-Paleczek, and Dávid Somfai Kara for their advice.