COMPENSATION AND KIN SELECTION IN THE LONG NINETEENTH CENTURY

: Using network models and quantitative methods, the present article provides a bird’s-eye-view of the Romanian novelistic translationscape published in volumes during the “long 19 th century”. The study approaches the cultural production of translated novels in the selected period from a relational perspective, aiming to investigate the connections between different publishers, with their respective editorial practices, and the translated authors selected from both major and minor source cultures. With this in mind, our paper will attempt not only to analyze the actor-network aspect of the translational networks established in the country, but also to provide an interpretive model for the selection of specific translated authors over others and their role in the cultural and nation-building process of early-modern Romanian culture.

serve very well the efforts to understand the literary cultures of the European peripheries and semiperipheries. Incomparable to the literary markets of France, England, or Italy, the Romanian literary market of the long 19 th century made up for its diminutive proportions through translations. In a culture where most of the books that shaped the intellectual and literary culture were imports, a problem occurred, and starting with the 1840s several voices started to complain about the lack of original literary works (Terian,. However, only in the interwar period did the number of Romanian novels surpass the number of translations. It was only in 1932 that the Romanian novel witnessed its first consistent "rise" (Terian,. Until then, translations flooded the Romanian market and created around 70% of this small literary culture's novelistic book market. This is, as we stated a few times before, according to Sean Cotter, what makes a "minor Romania": not the "failed state or potentially great one, but a translated nation" (Cotter 2). What we aim to analyze in the following is the editorial decisions that shaped this "translated nation" with respect to the novel and how the Romanian culture created "networks of authors" between 1800 and 1918 -a "long 19 th century". This kind of analysis was never conducted from a quantitative point of view until now, and we believe that the contemporary possibilities of visualizing editorial networks through quantitative and digital techniques are more than useful in understanding modern cultural spaces.
Moreover, in order to make sense of our data and networks, we thought of searching for a proper metaphor, borrowing from the natural sciences, specifically evolutionary biology. We will thus discuss the making of the editorial network by implying a strange yet -we believe -useful metaphor: the kin selection process. METACRITIC JOURNAL FOR COMPARATIVE STUDIES AND THEORY 8.1 218 Although in the last years several articles put forward interpretations of the modern translationscapes in Romania -where a translationscape is the "the sum total of texts of a given literary set visible in another" (Smith 751) -, none of them took into account the editorial process that translations imply. However, each publication of a foreign novel in translation is the result of a complex curatorial process, through which decision makers and cultural agents express both an ideological and a strategic program. Be it educationally motivated, ideologically claimed, aesthetically justified, or market-driven, choosing a novel to be translated into a peripheral or semiperipheral language and culture is a very important public decision. This is because, while many intellectuals in modern peripheries and semiperipheries can read novels in their original language, the general public cannot read French, German, or English. Yet our period represents the birth of a public national taste, where novels become the main source for public education and entertainment and their translation a form of adaptation to the modern world. This is where we would like to go back to Moretti's idea: "when a library confines itself to the canon […] it becomes scholastic, sterile; the more so, if it also excludes foreign novels" (Moretti 148). How canonical is the world translationscape in modern Romania? Moreover, what were the strategies through which the editorial decision broke this canonical aspect of world literature in a small modern European culture? And, finally, who were the actors of this network and how related were they to one another?
With these aspects in mind, we attempted to establish to what extent a network model could be applied to the Romanian production of translated novels in the long nineteenth century. The analysis of translations as networks of individuals, institutions or editorial entities is not exactly a novelty. Sociology of literature, with leading figures in the field such as Robert Escarpit, Lucien Goldmann or Pierre Bourdieu, has developed a series of analytical models that, especially in Bourdieu's case, have proven to be especially viable within the frameworks of World Literature studies. In addition to these traditional approaches, we propose the use of empirical data in the analysis of the Romanian translationscape. The ability to visualize patterns within an entire segment of cultural production opens up new perspectives through which we can question the complex networks that constituted the foundation of literary modernity in the country.
While data visualization cannot substitute actual interpretation, it can hardly be disputed that such approaches add to a rich hermeneutical and metadiscursive tradition by offering factual evidence of a "literary economy" (Jockers 28) that was previously approached intuitively rather than empirically. What we intend to do is employ network abstractions in order to visualize, within what was recently labeled "big translation history" (Roig-Sanz and Fólica), patterns that could serve as arguments towards an alternative approach to the study of literary translations.

Translationscapes in Competition
The Romanian literary culture, like many world cultures and most of the European cultures, owes much of its modern development to the French literary culture. In the context of the cultural dominance coming from Western culture, which came as a placeholder for the lack of consistent nation-building projects within the smaller, emerging nations, translations were seen as an integral part of the national cultural awakening, with writers "taking a pen in their hand, inspired by patriotism to translate" 1 (Pleșoianu, qtd. in Cornea 392-3). As Cornea put it, The considerable proliferation of translations in the fourth decade [of the 19 th century] could be explained through the optics of the cultural promoters of the time (Heliade, Asachi), through the prevalence of the Enlightenment's conception of culture, which emphasized the ethos of instruction and the assimilation of foreign experiences, especially as a preliminary step to affirming the autochthonous spirit. For Heliade and Asachi, translations contributed to a country's entry into Europe, to the purging of its mores, to the cultivation of taste, to the tidying and enrichening of the language (Cornea 395) 2 .
This dependency was discussed in numerous articles and essays and was often deemed as a self-colonizing complex. However, at the end of the 19 th century, the French model was already heavily disputed (Drace-Francis 114) through several strategies, described by Andrei Terian through the concepts of compensation, detour, and dumping: "compensation" can be defined as an "attempt to trim the heavy influence of a foreign literature by reorientation toward another foreign literature"; "cultural dumping" is the "multiplication of imports from as many cultures as possible, which should thus cancel the main dependence on German or French"; "detour" stands for the strategy "to retrieve the process of the systemic development of another peripheral literature that, in the meantime, has managed to integrate in world literatures" (Terian, National Literature 9). By following the "laws of imitation" (Tarde) in its process of "synchronization" (Lovinescu), the Romanian culture found ways to escape the selfascribed role of a "colony of the French culture" (Fundoianu 25)  Nordic, and the Asian are "the atomizers" (from 6 to 43 novels translated for each national culture or group): The table shows the division of non-French translations into Romanian on two levels of relevance: 1) the golden circle comprises German, English, Russian, and Italian novels (for the purpose of this study, I will only comment on the evolution of German and Russian translations), which will be called competitors. Not because they stood a chance against French novel translations, but because they fought each other for a place in Romanian culture (a funny way of putting this, of course, since no literature was actually fighting to penetrate Romanian culture at that point in history). The American novel emerges as a newcomer, embraced by local intellectuals even in the more conservative areas (influenced by trends like 'sămănătorism' or 'poporanism') especially for the sake of its novelty.
Somehow (and I will give a more thorough explanation of this), Romanian writers and editors felt that America was a young territory still under development, and viewed it as being equal to Romanian culture at best (no matter how strange it seems today); 2) the atomizers are primarily East European cultures (Hungarian, Czech, Polish, Serbian novels), Nordic (the Danish, Swedish, Islandic, and Norwegian novel), Asian (Japanese, Korean, Indian), alongside two literatures that -for some reason, either geographic distance or imperial proximity -never managed to enter the first category: the Spanish novel and the Austrian one" (Baghiu, Translations of Novels 95-6).

French (!Fr) novels (Baghiu, Translations of Novels 96).
This situation was caused by a specific chronology. As Graph.1 shows, it is only between 1900 and 1914 (before the fall of the editorial market during WWI) that non-French novels became -together, through "collaboration" -a serious competitor for the French novel in translation. A century of French domination through translations of novels was overturned by two decades of compensation and dumping, as Terian puts it.
We clustered a network of nodes that correlate data through the country of origin of the translated novels. Publishers are clustered here according to the novels translated and their respective country of origin. We separated the editorial groups that published  Figure 1.2.). This means that the same editorial groups that translate French novels translate many of the Spanish authors' novels. They are both also very close to the Universul publishing house.
Universul is at the same time far away from Russian, German, or Italian literature.
Meaning that the Spanish novel, an "atomizer" (a very poorly represented novel in Romanian translations during the long 19 th century) comes through the same infrastructure as the French "monopolist". The same thing happens to the Norwegian novel or to the Japanese novel. They are "closer" to the French cluster than to any other cluster, meaning that there were small literatures translated through the channels that brought in the French novel -and not otherwise, which is a very important detail.

Emphasis
At the same time, the network shows something else: the Polish, the Czech, and the Austrian clusters are really close to the American and English clusters. Meaning that the Central European, more "superior literary cultures" -bigger, older, more diverse than the Romanian one -were brought in through the channels that updated the translationscape at the end of the 19 th century. Finally, the greatest distance to the French cluster (The German, the Italian, and the Russian clusters) sheds light on another phenomenon. In a time of Austro-Hungarian dualism as Central and East-European imperialism (with a large part of the Romanian publishing system from Transylvania relying heavily on imperial administration), the Czech and Austrian novel came along with the editorial groups that brought the American novel. However, the Hungarian cluster is really close to the German, meaning that there was no coherent imperial editorial program. Moreover, the Serbian cluster is really close to the Russian cluster, meaning that the editorial groups that translated Serbian novels also had a great interest in translating Russian literature. Let us put this on a mental map: The Spanish, Norwegian, and Japanese novels are mainly translated by the editorial groups specialized in French translations; the Austrian, Polish, and Czech novels are mainly translated by the editorial groups specialized in American and English translations; the editorial groups that translate Hungarian and Serbian novels are mainly specialized in German and Russian translation.

Kin Selection in Modern Editorial Markets
The logic of this closeness, we believe, can be speculated around kinship: the French novel editors need Spanish literature for its prestige, for it to carry on its prestigious legacy. The first novel translated into Romanian was Spanish. It was Baltasar Gracián Y In order to answer the questions raised by the strange editorial connections in their relation to source cultures, we created a large network of all the published novels in translation between 1820 and 1918. We used a simple yet efficient community detection algorithm (modularity) provided by Gephi, and clustered the data extracted from The Chronological Dictionary of the Translated Novel (Istrate et al). Modularity places individual nodes into an aggregated group or cluster based on shared characteristics (Cherven 197). It looks for actors (i.e., authors and publishing houses) that are more densely connected to one another than to the rest of the network, unfolding "a complete hierarchical community structure for the network" (Blondel et al 10). Using our dataset, we grouped all the novels there (approximately 600 translated 226 volumes) according to some basic criteria: mainly the translated author and the publishing house that curated and published the translation. Figure 2 resulted from this process.
At first glance, the image might strike as odd. It groups together authors and publishing houses in a way that does not reflect a distinct translational topology. The sizes of the nodes (and their respective labels) are determined by the weight they accumulate (i.e., the number of translations made by a publishing house, or the number of translated texts authored by a single writer). In isolated cases, the strength of the relationships between nodes can be seen by the thickness of the edges. In this exhaustive, all-encompassing, democratic version, the network is non-binary and nonhierarchical; it reflects quantities and some degree of connectedness, but it requires several additional levels of data reconfiguration. In this mixed visualization of publishing houses and translated authors, we found that the clusters, formed according to the frequency of translations (of authors and by publishing houses), resemble a type of relationality comparable to the phenomenon theorized as kin selection. This is how we look at publishers: as strategic curators that try to secure "the survival of close relatives" of several literatures and canonical authors often by neglecting the evolution of the big names. In this process, some of the main authors who were translated can be seen as displaying a kin selection behavior for the survival of the group they belong to. Authors often behave -from this perspective -against their own well-being and advantageous position in the literary field, passing on the reproductive responsibilities of literary forms and techniques to their closest relatives within the network.
What we intended to showcase in this experiment were the merits of network analysis on a small scale, a localized cultural field in its interaction with foreign culture.
The resulting networks, we believe, provide crucial data needed for a relational approach to literary sociology . In this sense, Mustafa Emirbayer makes a very strong case for a shift from an inter-actional model of literary history, which describes interactions produced between independent entities, towards a transactional model, according to which all transactions between elements of a network can be patterned collectively, revealing relational communities. This, in turn, is crucial -in our case -to the understanding of the country's initial nation-building attempts to produce a culture through translations. In this specific context, publishers played an entirely different role in the curation of authors and works. Beyond playing the role of participants in an emergent book market, the publishers of 19 th century Romania were responsible for the careful selection of authors who could ensure a viable primary canonical collection that ultimately shaped the development of the early-modern Romanian novel.