The Global Network of Liberty: Toward a New Framework for Understanding the History of Political Concepts

Abstract

This article contends that liberty was already a globally connected concept during the late Middle Ages, and the Euro-American Enlightenment conception of liberty was only one of the many products of the global medieval legacies. Developing a network approach to concepts and applying it to primary sources in ten languages across Afro-Eurasia, I map how thinkers from different parts of the world contributed to the formation of the network. Recognizing this global network of liberty allows researchers to rediscover overlooked conceptualizations of liberty—as evidenced by examples of the Mongol Empire and its translingual politics with/in Europe, Persia, and China. Once innovations in specific contexts are placed back into the global network, revealed are global patterns of valorizing liberty, considering it either essential to possess or warranting caution. Both the findings and methodologies presented here prompt scholars to revisit the foundations of modern political thought from a global standpoint.

INTRODUCTION

This article proposes a new framework for understanding the history of the concept of liberty—broadly and tentatively understood here as the state of being able to act on one’s own will without significant external constraints (Podoksik 2010, 240). ¹ I contend that during the late Middle Ages (circa 1000–1600) (Hansen 2020; May 2012), ² liberty thus defined became a global concept in the following sense: Emic terms used across Afro-Eurasia to denote liberty were interconnected through various translingual practices resulting from multilingual governance, courtly encounters, and the spread of religions. Words for liberty in different languages paralleled each other in bilingual legal documents, resonated with each other in different translations of religious texts, and stood next to each other in bureaucratic glossaries and dictionaries. From Aristotle’s works to bureaucratic manuals, from Avestan scriptures in Sanskrit to Buddhist sutras in Tibetan, from contracts of south India to decrees of the Mongol Empire, different understandings of liberty, despite their different connotations, formed a conceptual network on a global scale. Thus, the concept(s) of liberty, instead of being a product of the West, is more accurately seen as a result of interactions among different parts of the world; the globalization of liberty did not start with the global expansion of the Euro-American empires but has a rich and complex history that predates them.

Recognizing the formation and transformation of the global network of liberty is important and urgent for at least two reasons. First, from a conceptual standpoint, it serves to provincialize the prevailing conception of liberty often associated with the Euro-American “Enlightenment.” This conception of liberty, as Annelien de Dijn recently characterizes it, “depends on the limitation of state power” and implies the protection of individuals’ lives from “government encroachment” (2020, 2, 4). While scholars have long scrutinized how this notion evolved from the Graeco-Roman tradition and laid the foundation for modern democratic societies (Constant 1820, 238; Edelstein and Straumann 2023; Schmidtz and Brennan 2010), my proposal is that this Enlightenment notion of liberty represents only one of the many outcomes of the medieval global network. This network includes at least two ideal-type approaches to valorizing this concept. The positive valorization, as I shall call it, contends that it is important for members of a political community to enjoy the right to act on their own will and be responsible for its consequences. The negative valorization posits that liberty is dangerous because it can lead to self-gratifying licenses, arbitrary actions, or autocratic measures, especially when enjoyed by those in power. Each approach—the positive or negative valorization of liberty—aligns with the Enlightenment concept in limiting the power of autocrats and protecting the lives of individuals in certain ways; however, the medieval approaches carry on further connotations, including the legal liability of those who acted freely, the public responsibility of officials who can act according to their wills, and the danger of arbitrary action of those who enjoyed freedom. The myriad explorations embedded in the medieval network thus reveal that the Enlightenment concept of liberty has flattened, marginalized, if not obscured, a variety of connotations that were closely associated with concepts of liberty as developed during medieval globalization.

Second, the excavation of the medieval conceptual network calls for a global turn in understanding the history of political concepts. It allows us to appreciate how the global expansion of Euro-American colonial empires, rather than delivering liberty to the rest of the world as claimed by the rhetoricians of these empires (see Abu-‘Uksa 2016, 18; Armitage 2000; Immerman 2012), disrupted existing conceptual and intellectual connections related to the idea of liberty. In other words, I challenge the widely shared assumption that political ideas could not have been global in any serious sense before Euro-American colonialism. For instance, in her otherwise insightful new history of liberty, de Dijn claims that “the Western political tradition has had far more impact than other, comparable traditions” (2020, 8). Her statement reveals the supposition that although conceptions of liberty in other traditions are rich in themselves, each of them remains a local and isolated phenomenon, whereas the Western concept alone has resulted in a global impact. To give another example, in their praiseworthy efforts to offer a trans-Asian history of freedom, David Kelly and Anthony Reid propose that “authentic conceptions of freedom are found throughout the Asian region” (Kelly and Reid 1998, back cover). This thesis is unfortunately compromised by Kelly’s insistence that these local traditions of liberty only formed the indigenous conditions in which “Western ideas of freedom have become widely accepted in Asia” (1998, 6). Again, the underlying supposition is that while concepts of liberty had been developed in many contexts, the Euro-Americans transformed theirs into a transculturally shared normative concept capable of grounding modern political life.

These two examples efficiently highlight a deeper problem in the study of political theory. Although Uday Singh Mehta has long posited that “liberal theoretical claims typically tend to be transhistorical, transcultural, and most certainly transracial” (1999, 51), the prevailing approaches to the key concepts and ideas are still premised upon a West-rest “dichotomy,” to use Loubna El Amine’s (2016) terms. This dichotomy prioritizes Western conceptualizations as “the foundations of modern political thought” (Skinner 1997; Ryan 2012), asserting that, due to the history of the past two or three centuries, it has already formed a part of the globally shared institutional and normative conditions (El Amine 2016; 2021; cf. de Dijin 2020; Kelly and Reid 1998). It then views the counterparts in non-Western traditions as alternatives that could potentially enrich Western understandings of the concept or facilitate the localization of the Western concept in the rest of the world.

Already, scholars have been questioning this dichotomy from different perspectives. Postcolonial thinkers have analyzed how imperialism and colonialism shaped the formation and globalization of Western ideas during the past three centuries (e.g., Abu-‘Uksa 2016; Mehta 1999). Comparative political theorists have demonstrated how intellectual legacies from different regions or traditions, as developed before European hegemony, can be reconstructed as normative foundations of modern political lives (e.g., Brown 2011; Crone 2013; Hashemi 2019; Jenco, Idris, and Thomas 2019; Kim 2008). Complementing these bodies of scholarship, I propose to examine the global formation of political concepts during the critical centuries when the world was increasingly connected, yet Euro-American empires had not yet dominated the rest of the world.

This global approach to the conceptual world of the late Middle Ages has unique merits. As Nazmul Sultan has pointed out, the existing scholarship of “empire and anticolonialism posed questions pertaining to the applicability of ideas and norms” developed in the Euro-American context but has not delved into “the foundational matters in the discipline of political theory” (2024, 10). Tracing the evolution of the global network of liberty prompts us to rethink the historical processes through which foundational values in political theory emerged. Once we recognize how the global conceptual network emerged in the first place, we can subsequently explore different traditions through the lens of connection, investigate the innovations of each individual region, and rethink how Euro-American empires appropriated or undermined the global intellectual legacies of premodern times. What we will learn, accordingly, is not merely how our current ways of thinking about liberty “reflect a series of choices made at different times between different possible worlds,” in Quentin Skinner’s terms (1998, 117, emphasis mine); even more importantly, we will be able to engage with the sophisticated conceptualizations that have already emerged during the historical encounters of different historical worlds.

The global turn in the study of medieval political ideas is indebted to, and necessitates, closer cooperation between the scholarship of the “Global Middle Ages” and political theory. In 2012, Jennifer Pitts pointed out that in contrast to the fields such as anthropology and world history that have “produced particularly sophisticated accounts of the asymmetrical interactions that brought about global modernity,” “political theory has come slowly and late to” the study of empires and their global expansion (Pitts 2012, 353, 352). The same is true today in the sense that, relative to anthropologists and historians, scholars of political thought are not used to thinking of the medieval world as a connected one via variegated commercial ties, pilgrimage routes, the spread of religions, and the expansion of empires. By comparison, over the past decades, medieval historians have detailed the processes by which the world became increasingly connected (Hansen 2020; Northrup 2005; Subrahmanyam 2012). Economists, archaeologists, and scholars of religion, science, and literature have likewise traced the evolving networks of goods, objects, texts, and knowledge across Afro-Eurasia during the Middle Ages (e.g., Abu-Lughod 1995; Blake 2016; Golden 2000; Pollock 2009; Zarakol 2022). The rapid development in adjacent fields suggests that it is promising for political theorists to delve into premodern contact zones and examine discursive practices.

Nevertheless, researchers of political thought face unique challenges when it comes to transcultural interactions (Moyn 2012; Pernau 2016; Pocock 2019). Notably, toolkits for identifying, reconstructing, and interpreting conceptual connections remain rather limited. For instance, the fact that Persian and Chinese words related to liberty were glossed together in one compilation by no means implies that they are commensurable with one another (see Jenco 2015, 17). What connects different words denoting different concepts of liberty is usually everyday use loaded with ambiguity and contradiction, if not misunderstandings and mistranslations. Further, connections during medieval globalization were embedded in contexts characterized by the asymmetry of power, including clergy seeking patronage or protection or petitioners seeking justice. A study of the global connections of concepts, including their formation and evolution, cannot be based on any romanticization of the pre-European hegemony world as a coercion-free environment. Simultaneously, although this study owes a tremendous debt to centuries of philological and historical research, it must make an interpretative leap to show how the translingual practices of words and concepts have raised theoretically provocative questions in unique ways. In summary, a new set of methodologies is needed.

Against this backdrop, I start by elaborating on the network approach by taking insights from fields including network analysis, translingual practice, and global conceptual history (Latour 2007; Liu 1995; Pernau and Sachsenmaier 2016). Applying this method to a large group of primary sources in ten languages across Afro-Eurasia, all of which I read in their original languages, I depict a significant portion of the global network of liberty during medieval times. Recognizing this global network of liberty allows researchers to take two additional steps. First, it enables scholars to identify critical nodes (or clusters) of concepts that have long been—and would otherwise be—overlooked. I shall argue that liberty, as developed through the translingual practices under the Mongol Empire, was one such example (Brook, van Walt van Praag, and Boltjes 2018; Favereau 2021). Second, the network approach allows researchers to reappreciate the significance of an idea, not only in the context of its specific spatial and temporal origin (such as the Mongol Empire) but also through the lens of the globally connected valorization of liberty, both positive and negative.

In sum, focusing on the global network of liberty, I seek to exemplify the following approach to political concepts: the better we grasp the global network and its transformations, the better we can excavate and appreciate local contributions, and vice versa. The three sections below proceed in a way that showcases how to achieve this virtuous circle, with the hope that the method can be productively applied in future studies focusing on other periods or concepts.

CONNECTION FIRST: A NETWORK APPROACH TO CONCEPTS

When analyzing a political concept, researchers often start by clarifying its meaning, tracing the evolution of its various semantic contents across complex histories (Koselleck 2002, e.g., 82, 209). This “clarification first” approach is beneficial for various purposes, but at least in some cases, ambiguity is intrinsic to the deployed concept. For instance, as anthropologist Stephan Palmié has pointed out, victims of horrific events “actively analyze and render comprehensible their situations” in ways that reject the kind of clarification that philosophers pursue in their conceptual labs (2002, 18). It is often the case that the vagueness of concepts is not merely something to be overcome but rather something to be appreciated. “Ambiguity” is a part of “political and social convulsions,” as Reinhart Koselleck (2004, 45) has noted in his reflection upon “conceptual history;” it “performs political functions,” using the terms of J.G.A. Pocock’s (1971, 20), the celebrated historian of the Cambridge School.

The insights from conceptual history, the Cambridge School, and other disciplines lead to the further question of how to transform ambiguity itself into a productive site for political theoretical research. For this purpose, this section elaborates on the “connection first” or network approach. I propose that if we examine vague uses not in isolation but as interconnected acts between words and concepts, we cannot help but discern clusters of relationships. At this point, the vagueness that previously seemed like a dead end for research reemerges as a critical entry point to rich political theorization. I shall begin with a hypothetical example and then gradually unpack the merit and limitation of the “connection-first” approach.

Let us assume that a person writes the English word “freedom” underneath the French “liberté” to make sense of the latter. Clearly, this phenomenon does not imply an equivalence of semantic meaning between these two words. ³ Following the clarification-first approach, one needs to investigate the uses of these two concepts in each language and in this contact zone (Koselleck 2002, 27). The network approach, in contrast, underscores the fact that because of this person’s act, a link—or an edge—emerges between these two words (cum nodes). This edge thus denotes that in a given context, a person uses one word to make sense of the other. This approach, which treats “freedom” and “liberté” as linked nodes, at least temporarily sidesteps the daunting question of the extent to which the semantic or pragmatic meanings of “freedom” and “liberté” align. Instead, it prompts scholars to trace how concepts in different languages are interconnected to each other through concrete acts of translingual practices.

A particular advantage of this approach is that it reveals the centrality of certain nodes, inviting researchers to think about why these emerge and what that emergence means. For instance, if another person writes the English word “freedom” underneath the Latin “libertas,” a three-node network thus emerges, with the English one as the center of this network. This prepares the ground for the critical question of what it means that the English term “freedom,” as well as the totality of historical processes associated with this term, mediates the other two words. If a key insight of Koselleck and conceptual history is that concepts are both indicators of change and forces behind historical changes (Koselleck 2004, 43; see also Pernau 2016, 484), the network approach adds that the links between concepts are no less important as both indicators and forces. Just as Mark Granovetter (1983) revolutionized our understanding of social interaction by highlighting the strength of weak ties, my proposal is to seriously consider conceptual links that appear weak and vague.

Thanks to the recent development of global intellectual and conceptual histories in recent decades, the question of whether and how one should “distinguish between successful and failed translation” has received increasing attention (Pernau and Sachsenmaier 2016, 18; see also Karl 2002; 2017). From the perspective of the network approach, the task is no longer to adjudicate the success of a translation, but rather to trace how a translational act transforms the network. Let us assume that another individual connects the English word “library” with the Latin “libertas.” In this case, the established semantic meanings of these words do not overlap, and a language teacher would argue that this student simply commits a mistake in thinking “libertas” means library. However, via the lens of the network approach, the edge between “library” and “libertas” transforms the network and deserves no less attention than the aforementioned edges.

The network approach to concepts has significant limitations. As with social network analysis, it flattens the differences between edges, viewing different connections as ontologically equivalent (Davison 2019). Any conceptual network is a simplified approximation of the underlying connections and cannot replace close reading that delves into the nature of the connection. It is therefore important to bear in mind that the network approach is developed to grasp the vague uses of concepts in contact zones; it is helpful to view it as a scaffold that enables researchers to grasp the global connection of concepts, identify critical clusters, and appreciate local innovations against the backdrop of global transformation. Once these goals are achieved, the scaffold can be abandoned.

What the network approach unveils, first and foremost, is that words and their related concepts were already interconnected well before modern times. Figure 1 depicts a network that I have generated based on a preliminary survey of glossaries, dictionaries, multilingual documents, and translated scriptures or treatises, most of which were compiled from about 1000 to 1600 (further explained in the following sections). Each node represents an emic term or phrase that either signifies acting on one’s own will (the working definition of liberty in this article) or a term used to translate or explain the former. Each edge thus indicates a translingual or hermeneutic practice in history, with the date indicated on the edge. Clearly, Figure 1 is not intended to be comprehensive. It features a portion of the network of liberty of medieval times (depending on the languages I have studied and the sources I have identified so far). Nevertheless, this portion alone covers the Afro-Eurasian continents and reveals the multi-centric nature of the global network of liberty as developed up to the end of the sixteenth century.

Figure 1. The Global Network of Liberty, ca. 1000–1600

Three clusters from Figure 1 command our attention. The first is the one at the bottom. It is well-known that in his Politics, Aristotle considers the understanding of liberty (eleutheria ἐλευθερία) as the ability to act as one wishes (bouletai βούλεταί) and disparages it as “licentious” and irresponsible (Edelstein and Straumann 2023, 1038); in his Nicomachean Ethics, Aristotle theorizes voluntary or free action as those following from wishes (boulēsis βούλησις) depending on oneself (eph’ hēmin ἐφ᾽ ἡμῖν) (Meyer 2014). While Aristotle’s ethical and political theories have been the subject of rigorous studies, it suffices here to note that the Greek notion of “wish” (boulēsis) mediates these two fields, and the Aristotelian legacies played a formative role in the making of medieval Islamic and Catholic philosophies. To give just one example, Averroes (1126–1198), in his Arabic commentary of Aristotle’s Metaphysics, uses the Arabic word “the object being willed” (al-murād المراد) to explain the Greek conception of the object being wished and theorize intentional action (Averroes 1964, 1592; Phillipson 2017, 5; Somma 2018, 139).

Central to the current purpose, the Arabic notion of “what is willed” gives rise to a critical word in Persian: khwud-murād خودمراد (self-willed), a Persian-Arabic compound combining the Persian word for “self” and Arabic word for “will.” The Persian notion of self-willed links the Graeco-Arabic cluster to the cluster of ideas developed in Mongolian and Chinese languages, which sits in the center of Figure 1. Under the Mongol Empire that ruled over the vast lands from East Europe to East Asia, the Persian(-Arabic) word khwud-murād was introduced into China and glossed with its Chinese counterpart, ziyou 自由 (following oneself or one’s own will), by translators working on diplomatic documents for Ming China (1368–1644). In fact, under Mongol rule, both Persian and Chinese terms were once used to interpret the Mongolian phrase öber-ün durabar (on one’s own will). This phrase specifically emphasizes free action’s association with legal liability (see below). The connection of the Persian, Chinese, and Mongolian terms, as I shall argue below, reveals how liberty emerged as a critical political and legal concept under the Mongol Empire, one that confirms the privilege of those included as agents of the empire while holding them liable for their actions.

Meanwhile, the Mongolian translation of the Buddhist sutra printed in Beijing in 1312 further employed the Mongol phrase öber-ün durabar to interpret the Tibetan and Sanskrit notions of self-independence (Tibetan: raṅ dbaṅ du རང་དབང་དུ་; Sanskrit: svādhīnavṛttayaḥ स्वाधीनवृत्तयः) (Schwartz 2018, 13). ⁴ In other words, the Tibetan concept connects the Mongolian, Persian, and Chinese theories of acting on one’s will to another cluster of concepts that had been developed in the Sanskrit cosmopolis and the Buddhist sphere—the one on the left of Figure 1. A closer examination of this connection will reveal how Buddhists transformed religious concepts into political ones during the Mongol Empire.

To further test the potential of the network approach, I add an additional cluster (with edges highlighted in blue and dates in red) to the global network of liberty of late medieval times (Figure 2). This additional cluster represents the reception of J. S. Mill’s On Liberty in languages and cultures related to those exposed above. It shows that the Mongolian, Hindi/Sanskrit terms, and Chinese terms that already formed part of the medieval global network of liberty reemerged as signs translating the English conception of liberty, as developed in J. S. Mill’s On Liberty; these terms continued to play an important role in the (trans)formation of the global network of liberty during the late nineteenth and early twentieth centuries (Abu-‘Uksa 2016; Berest 2019; Fung 2010). In other words, Figure 2 reveals that the global network of liberty from medieval times and that of modern times are directly connected. The implication for future research is two-fold. On the one hand, it is imperative to think of the formation of modern conceptions of liberty as evolving from a global legacy, rather than solely from the European Enlightenment. On the other hand, due to the increasing power imbalance between the West and the rest, nineteenth-century intellectuals from the non-Western world ended up reformulating indigenous notions of liberty based on their reception of the theories of J.S. Mill and Jean-Jacques Rousseau, among other Western thinkers (Charfi 2005; Fung 2010; McMahon 2012).

Figure 2. The Global Network of Liberty, ca. 1000–1600 and ca. 1850–1910

While further research is needed to test and develop these observations, Figure 1 draws our attention to the Mongolian term, which forms a key nexus of the global network. At this point, questions arise about what this term meant and how actors used it to do things in context. The resulting case study, pursued in the next section, will reveal how the global network can give visibility to long-overlooked conceptualization and theorization of particular time and space.

ZOOMING IN ON ONE NODE: LIBERTY AND LIABILITY UNDER THE MONGOL EMPIRE

A common misconception about liberty is that Western cultures are exceptional in considering it a particularly important value (Strauss and Cropsey 1987, 908). This zooms in on one node of the global network and argues that the Mongols laid particular emphasis upon the notion of acting “on one’s own will” (ö’er-ün dura-‘ar), at least since the rise of Činggis (Chinggis) Khan, ca. 1162–1227). This Mongolian phrase embodies two interrelated yet distinct connotations. The first suggests one enjoys the privilege of acting in the absence of restraints or commands of someone else, of pursuing one’s action as one wishes. The second suggests that a person who acts “on their own will” is legally liable for their actions because they are following their own deliberation and not under the control of someone else. An appreciation of this Mongol phrase will show how liberty played an important role in a hierarchical society—contra the prevailing assumption that pits it against hierarchy—and will serve as an entry point to the richness of the global network in general (see the next section). I proceed through a close reading of three kinds of sources: the narrative The Secret History of the Mongols, the decrees of different parts of the Mongol Empire, and Buddhist translations produced under the Mongol Empire.

Both connotations of “on one’s own will,” as a privilege and as an entailment of legal liability, can be found in The Secret History of the Mongols (hereafter the Secret History), the Mongolian historical narrative finalized around 1252 and often regarded as the single most important account of the rise of the Mongol Empire (Atwood 2007). Concerning the first connotation of acting without restraints, a notable example comes from a 1203 decree of Činggis Khan. That year, Činggis Khan defeated a key rival of his and captured an enemy general named Qadaq Ba’atur. After pardoning Qadaq Ba’atur from execution, the khan made Qadaq Ba’atur and his followers the hereditary bondservants of one of the Khan’s loyal subordinates named Quyildar (Atwood 2023, 252n73). This was intended as punishment for the former and reward for the latter. Thus, the sons of Qadaq Ba’atur were to continue serving those of Quyildar. Further, the decree stated:

When girls are born [from Qadaq Ba’atur and his followers], do not let their father and mother marry them on their own will. ⁵

Accordingly, the khan deprived Qadaq Ba’atur and his followers of an important privilege in Mongol society,—that is, the ability to marry their daughters according to their will; instead, these daughters were obliged to serve the daughters of Quyildar. The example vividly illustrates that the Mongol Empire (ulus) was a hierarchical order in which those deprived of liberty served those who had this privilege. It also reminds us that liberty, while often viewed as a key value in egalitarian societies, can also play a critical role in maintaining hierarchical and monarchical ones.

When the phrase “on one’s own will” appears again in the Secret History, it exemplifies the second connotation, namely, the entailment of legal responsibility. In 1219, Činggis Khan started campaigning against the Khwarazmian empire in Central Asia. One of his generals, Toqucar, decided to pillage the cities of a warlord named Qan-Melig, prompting the latter to side with the Khwarazmians, the enemy of Činggis Khan. In punishing Toqucar, Činggis Khan established a judgment (jasaq) that would “serve as a precedent in similar cases” (Atwood 2023, 284). The decree reads:

Toqucar, on his own will, captured the frontier cities of Qan-Melig and turned Qan-Melig into an enemy. I make it a binding precedent and behead him. ⁶

Precisely because Toqucar pillaged the city on his own will, he alone was accountable for this misbehavior and was subject to punishment.

Thus we see that during the early thirteenth century, ö’er-ün dura-‘ar, carrying with it the dual connotations of acting without restraints but with legal responsibilities, was already an important legal concept for the Mongols. It is worth stressing that both examples appear in Činggis Khan’s decree, which lays the legal and normative foundation for the Mongol Empire governed by his successors. During the thirteenth and fourteenth centuries, the Secret History, as well as decrees and legal codes promulgated by Činggis Khan, continued to play a prominent role in different parts of the Mongol Empire. This legacy was particularly important in two constitutive regimes of the Mongol world: the Ilkhanate ruling over Persia and the Yuan governing China and its surrounding regions. To further appreciate the Mongolian phrase “ö’er-ün dura-‘ar” and its two usages, I introduce three documents as examples: (1) Ilkhan Argun’s letter to Pope Nicolai IV of 1290; (2) an Ilkhanate decree of 1320; and (3) a decree of Yuan China, issued in 1291 and republished in 1323. Together, they offer three snapshots of how the Mongol ruler used the concept of liberty in the Mongol-Europe, Mongol-Persia, and Mongol-China contact zones.

In 1290, the Ilkhanate and various European power holders were exploring the possibility of an alliance against the Muslims in the Middle East. Against this backdrop, Pope Nicholas IV sent a letter to Arghun, the Mongol Khan of Persia, claiming that Christianity was the true religion and urging him to convert. Arghun refused with the following words:

We say, as descendants of Činggis Khan, our own Mongol people, being of our own will, whether some of us adopt silam (namely, Christianity) and some do not, only the Eternal Heaven knows. ⁷

Linguistically, the letter first characterizes “our Mongol people” (öber-ün Mongγolǰin)—or “we proper Mongols” (Mostaert and Cleaves 1952, 450), and then stresses that this collective group was free in choosing which religion to follow. The phrase öber-ün … durabar is, first of all, a reconfirmation of the privilege that the Mongol people enjoyed, a usage that already appeared in the Secret History. Its purpose is less about conceptualizing a form of self-determination, as it was in the nineteenth or twentieth centuries, but rather to serve a concrete political purpose. By highlighting that different Mongols would make their own choices, the khan left open the possibility of being converted and forging a military alliance. Last but not least, the decree insisted that the Eternal Heaven—or mongke tngri—the highest deity of the Mongol tradition is the ultimate authority for determining whether certain members of the Mongol people will endorse Christianity and whether this was a good choice.

This letter, having survived in the Vatican Apostolic Archive, must have been translated into Latin during the early thirteenth century (Fiaschetti 2020), although the Latin copy remains to be found. For the current purpose, it suffices to emphasize that the liberty a group of people enjoys to decide on their religion had already entered the Latinate sphere from the Mongol tradition. In other words, while Figure 1 shows that medieval terms in the documents we have today form an extensive network, it is worth stressing that many connections and their sources have not survived. Against this backdrop, even a single node itself may imply the existence of a network behind—as the details of other nodes are no longer known to us.

The second document was issued by Busayid Baghatur Khan, a grandson and successor of Argun. Dated in 1320, the decree sought to settle a dispute of succession between two brothers from the Persian-speaking Kurds in the Caspian region. The disputants were, as Vladimir Minorsky (1954) has determined, the brothers-in-law of Safi al-din Ardabili, the renowned reformer of the Safavid and an ancestor of the Safavid dynasty of Iran (1501–1736). The document is of tremendous value in understanding the early history of this ruling house in the history of Iran and illustrates how the Mongol court interacted with and ruled over local religious communities. The decree begins by stating that a man named Badaradin Abul Maqmad had come to the court and presented a petition (Cleaves 1953, 16). After citing this petition in detail, the decree proceeds to command the Samsadin Maqmad, the wrongdoing brother, to return what he had misappropriated. This part of the decree is translated below:

This Elder, Badaradin Abul Maqmad Maqmud, having given prayers, came [to us] and made the petition, which says:

By the decree, in my father Elder Čamaladin’s place, [I was] to sit on the prayer mat of Elder Ibrayim the martyr and be appointed over the endowment of the property. However, my brother Šamsadin Maqmad, not letting me approach, seized the decrees and documents and ruined the endowment.

If this is true, how has he, on his own will and in this manner, committed these illegal acts? ⁸

In his meticulous study, Francis Woodman Cleaves, a grand scholar of Mongolian studies, points out that this “öber-ün durabar” is exactly what appeared in the Secret History. My translation further stresses that “on his own will” and “in this manner” (eyimü) are juxtaposed to set the stage for the decision that any misconduct he had committed should be reversed. By deploying the term “öber-ün durabar,” the khan held the wrongdoer accountable for his actions by stressing that he had acted as such according to his own will. In other words, precisely because the wrongdoer had used and abused his privilege to act freely, he would be liable for his actions. In this case, while the Mongolian version had probably been translated into Persian immediately, the Persian version has not survived—and concrete examples of the Mongolian-Persian connection will be discussed in the next section. ⁹

The third example is from the Mongol Yuan Empire in China. After conquering Jiangnan (roughly the Lower Yangzi Delta) in 1276, the Mongol Yuan enlisted various captured craftsmen into different governmental bureaus. Supposedly, they produced a specific number of artisanal goods for these bureaus and enjoyed tax remissions. The problem was that the powerful officials and magnates of the empire forced these governmental craftsmen to work for them. Against this backdrop, in 1291, the Secretariat, the central governmental institution assisting the Mongol khan in routine administration, received a petition reporting such misconduct. It forwarded the petition to the Khan with a recommended solution: to punish the wrongdoers according to a previous decree.

Qubilai Khan replied with a rescript. He stressed that the issue at hand was more complicated than officials and magnates misappropriating artisans, as his Secretariat officials had memorialized; ¹⁰ rather, it concerned the tyranny of Samgha, the financial minister from “a partly Uighurized family” of Tibet (Petech 1980, 195). Samgha was responsible for collecting revenues for the khan from 1287 until the beginning of 1291, when he was declared guilty of abusing power for his own purposes and executed. According to the khan’s rescript, the problem was that Samgha and his followers, denounced as “thieves,” had illicitly forced the artisans to labor for their benefit. Against this backdrop, Qubilai Khan singled out another group. They had not submitted themselves to Samgha; they were “acting on their own will” (ziyoude) not Samgha’s. ¹¹ Although they had misappropriated artisans, they were not to be labeled and captured as “the thieves” following Samgha; for this reason, other punishments (supposedly lighter ones) would follow. Essentially, given that it had only been nine months since Samgha’s fall from power, the Khan prioritized purging the followers of Samgha from the Jiangnan government over addressing the generic issue that the magnates and officials misappropriated governmental artisans.

Qubilai Khan’s rescript survived in the form of the hybrid language known as Sino-Mongolian, which involves writing down the spoken Mongolian language exclusively using Chinese characters (Birge 2017, 3). Some Chinese characters are used as syllabograms to represent Mongolian morphological changes, particles, and nouns, while other Chinese characters stand for words in colloquial Chinese language(s). In this hybrid linguistic environment, the colloquial Chinese word ziyoude 自由的 ([one] who follows one’s will) appeared as an adjectival word, describing the persons who acted according to their own will. ¹² Here, as in the example above, the Khan employed the term “acting on one’s own will” to identify a specific action of a given person that made that person liable for punishment. The theoretical problematic here is not the metaphysical-moral one about whether a person possesses free will and is morally responsible for their actions (although this problematic did appear in Figure 1 with the nodes connected to Aristotle’s term for rational wish [boulēsis]). Instead, it is a political-legal problem that concerns, first and foremost, what kind of legal decision is appropriate when their actions result in a legal offense or other unfavorable outcome.

The three documents from 1290 to 1323 confirmed, historically, that the notion of acting on one’s own will held significance across the Mongol Empire—temporally from the rise of Činggis Khan to the early fourteenth-century successors and spatially from Persia to China. This notion entered the translingual practice of the global fourteenth century, interacting with its counterparts in Latin, Persian, and Chinese. Given that all the documents above come from the words of Mongol khans, the critical question concerns whether other actors deployed this notion for their own purposes and how they further theorized this notion. An important example is Čosgi Odsir’s Mongolian translation of a Buddhist text published around 1312, namely, the Bodhisattvacaryāvatāra, or A Guide to the Bodhisattva’s Way of Life (Cleaves 1954, 13; see Supplementary Appendix A). ¹³ In the conceptual universe of the Mongol Empire, liberty is the privilege to do things as one wishes, whether claiming a piece of property, marrying one’s daughter to build a family alliance, or attacking a city to advance military goals. But what if one’s actions affect others who might claim the same privilege, such as claiming the same piece of property? Čosgi Odsir argues that one rightfully enjoys (edleküi) the privilege to act on one’s will when there is no other contender. This happens not when all contenders are physically defeated but when one achieves a stage of enlightenment. As one joins the community of the enlightened, one transcends the realm where individuals contend with and harm each other and is thus in a position to enjoy. Čosgi Odsir’s argument can be reconstructed as follows: the idea of liberty as understood as a political privilege can be better (or truly) achieved within a different kind of political community (of the enlightened ones) whose members are no longer in a zero-sum competition for properties and privileges.

In summary, this section demonstrates how an understanding of the global network, as visualized in Figure 1, informs case studies and leads to the discovery of previously unknown traditions. As I have shown, no later than the Secret History, the Mongols valorized liberty as a privilege that both individuals and groups could enjoy or lose. With the notion of “liberty,” the Mongol khans ascribed considerable agency to their privileged representatives across the empire, while holding them accountable for the actions they took at liberty. With Čosgi Odsir’s translingual practice between Mongolian and Tibetan, the question emerges: under what circumstances can one rightfully or truly enjoy one’s liberty as a privilege to act as one wishes? For Čosgi Odsir, one will be able to do so only in the community of the enlightened ones. During this process, the Mongol concept of liberty (or ö’er-ün dura-‘ar) emerged as a shared and interconnected notion across the vast regions of the Christian, Persianate, Sanskrit, and Sinitic spheres. The complicated relationship between liberty as something to enjoy and something entailing liability will further reveal rich conceptual transformations of the connected globe, as we shall see below.

ZOOMING OUT TO GLOBAL TRANSFORMATION: POSITIVE AND NEGATIVE VALORIZATION OF LIBERTY

In this section, I propose that the global network approach is more than a tool for visualizing connections and excavating ideas that have fallen victim to the “epistemicide” under Euro-American empires (de Sousa Santos 2014), as explored in previous sections. Instead, it embodies a deeper methodological contention. Complementing existing approaches that situate a concept in its immediate political contexts (such as Qubilai’s court) or a certain period of evolution (such as the Mongol Empire from the thirteenth to fourteenth centuries), it is no less productive to appreciate it as part of the broader horizon of globally connected transformation. Put differently, if the previous section shows how researchers can zoom in on a specific node of the visualized network (Figure 1), what follows explores how to zoom out, re-embedding the Mongolian or other emic terms back into the broader picture. I will first follow the late medieval translators and elaborate on two ideal-type approaches to valorizing liberty: the negative valorization takes it as something detrimental to the political community, while the positive valorization considers it as something essential to members of the community. Then, I argue that the continuum between the negative and positive valorization of liberty allows us to reappreciate the theoretical significance of practice embedded in specific contexts of the late and post-medieval world.

In 1407, not long after the fall of the Mongol rule in East Asia, the Yongle emperor of Ming China established the Translation Bureau—known as the Bureau for Four Barbarians—in Beijing, the capital of the Mongol Yuan empire (Lotze 2016, 31; see also Liu 2008; Wuyun-Gaowa 2014). This move was part of Yongle’s broader vision: to project Ming power over Eurasia and establish the Ming Dynasty as the true successor to the Mongol Empire. The Bureau for Four Barbarians took on the critical responsibility of translating documents between Chinese and various languages—including Mongolian, Persian, Burmese, and Thai. As time went on, the officials and scholars of the institute compiled glossaries and bilingual texts to facilitate pedagogy and translation. These reference materials were recopied and further disseminated in subsequent centuries, resulting in the survival of multiple manuscripts. The translators working for the Chinese emperor glossed the Chinese word ziyou (following one’s own will) with the Persian khud-murād. This is evidenced by the manuscripts now housed at the Berlin State Library, believed to be late sixteenth-century handwritten copies (see Figure 3).

Figure 3. The Sino-Persian and Sino-Mongolian Glossaries of the Late Sixteenth Century

Note: Images in this figure are from the book Huayi yiyu (Ca.1579, 1.84v, 6.109v).

The connection between the Mongolian, Chinese, and Persian words leads to the clarifying question of what these terms mean. The Persian notion khud-murād (either as a two-word phrase or a compound word) is a combination of the Persian words khud (self) and murād (which means either a rational wish or its object as in Arabic). The latter, coming from the stem r-w-d, has a rich history in Arabic writing. Franz Rosenthal has shown that words derived from this stem already appeared frequently in Abū Ḥayyān at-Tawḥīdī’s (d. after 1009) philosophy of action. This late tenth-century thinker wrote, as Rosenthal translates, “the one who makes a choice (mukhtār) is the one who wants (murīd) the best (khayr) of two things, … without being in any way forced to make a choice” (2015, 42n49). Rosenthal argues that this ability to “want the best” is not to be conflated with the problematic of freedom of will (versus determinism), and is better understood as a theorization of the connection between “human action and human beings” (42). The combined Persian notion, meaning literally “self-willed,” can be found in a variety of medieval writings (Hakimazar 2009). To give an example, Amīr Khusruw (1253–1325) was a renowned poet of the Delhi Sultanate. One line from his love poem states:

Because of you, Khusruw is restless to be with you; what shall I do to the heart when it is self-willed? ¹⁴

Here, Khusruw describes a mental state where his heart is taking action on its own (to be with the loved one), and he is experiencing difficulty in controlling this will. What is at stake is not the weakness of the will, but rather the will’s takeover of the heart, disregarding further constraints. The political significance of this idea of self-willed would be further explored by later thinkers of the Delhi Sultanate. For instance, in 1510, Shaikh Meḥmood bin Shaikh Ẓiā’uddīn Aḥmad, a scholar working for the Sultan of Delhi, compiled a Persian-to-Persian lexicon titled Toḥfatus Sa’ādat (The Gift of Happiness) (Shaikh Meḥmood 2019). The lexicon explains the term khudkāmeh (one who imposes his will or autocrat) in the following way: “(the letter u [in khudkāmeh] being silent) self-willed [khud murād].” ¹⁵ Shaikh Meḥmood by no means intends to suggest that these two terms—“autocrat” and “self-willed”—are equivalent, nor did he seek to provide a full-fledged clarification of each, which would require an entire treatise. What it suggests is that if those in power follow only their own will and impose their will on others, it can lead to arbitrariness or autocracy. While this exemplifies the negative valorization of liberty, the glossary practice connects liberty and autocracy, reminding us of the complicated relationship between the two.

In the Sinitic sphere, the word ziyou can also imply a tendency to act arbitrarily. To provide just one example, in 1216, a minister named Chen Gui 陳規, when advising the Jurchen ruler he served, pointed out the political danger where “ministers in charge of state affairs act solely according to their own wills and wield power and privilege at their liberty (ziyou).” ¹⁶ Chen Gui, just like Shaikh Meḥmood, stressed that if those enjoying substantial political power act according to their own will, it leads to a kind of arbitrary rule that jeopardizes the polity. Simultaneously, ziyou, implying the ability to follow one’s will (as in Abū Ḥayyān at-Tawḥīdī’s theorization), embodied the positive valorization as well. Late thirteenth-century Chinese-language drama already employed the word ziyou to mean a privilege to act freely. ¹⁷ The Kingdom of Korea, where Classical Chinese was used as one of the lingua francas, provides another critical example. In 1482, a minister reminded the king of Chosŏn Korea, in Classical Chinese, that the country was previously subjugated under Mongol rule and lost its liberty:

The [Mongol] Yuan Empire established the executive officials [daruγači to oversee our state], not allowing anything to be done at the liberty (ziyou in Chinese pinyin or chayu 자유 in Korean hangeul) of our state. ¹⁸

Here, the meaning of ziyou/chayu involves protection from government interference, which aligns closely with the Enlightenment conception of liberty. Even more importantly, what should have been protected was not individual liberty but rather a country’s right to decide its own affairs.

At this point, it becomes evident that the cluster of Mongolian, Persian, and Chinese terms does not represent a single notion of liberty but points to at least two ideal-type valorizations, positive and negative, each leading to diverse and rich theorization on its own. The positive valorization stresses an individual’s or even a country’s right to pursue what they want, while the negative valorization reminds us that this pursuit, when conducted by someone in power, can lead to arbitrary actions resulting in harmful outcomes for others or for the polity.

The Mongolian legal notion of acting “on their own will” itself attests to the continuum. When it means the privilege of acting in the absence of restraints or commands from someone else, it suggests that liberty is an important privilege to have. When it stresses that a person who acts “on their own will” is legally liable for their actions, it implies that if one acts arbitrarily, it will lead to negative consequences for the community, for which punishment will follow. In other words, once one re-situates öber-ün durabar back in the global network, its theoretical significance reveals itself. It encapsulates the dynamic relationship between the positive and negative values of liberty in the following way: in a society where inequality prevails, precisely because liberty is valuable as a positive force, the privileged ones gain a position to enjoy it, and this enjoyment requires a conceptual framework to constrain actions at liberty.

The very continuum between positive and negative valorization provides a new way of thinking about the global transformation of liberty during medieval and post-medieval times (ca. 1600 to the present). “Liberty” in mid-seventeenth-century England and the European Enlightenment carried on the positive valorization out of the connected medieval legacies. As Skinner and others have demonstrated, those living through the Interregnum (1649–1660) in England pressed into the question of whether individuals can still enjoy liberty as a privilege or right to do things as they wish once they entered a political society under a government, prompting thinkers from Milton to Hobbes to rethink the proper form of government (Skinner 1998). It is worth stressing that Europe was not alone in furthering the positive valorization. To give only one example, during the mid-seventeenth century, the Manchus rose as the overlord of East Eurasia and established the Manchu Qing Empire, a post-Mongol regime that ruled over China, Mongolia, and Tibet until 1912. ¹⁹ To maintain the order in Mongolia that it had conquered, this Manchu Qing empire promulgated a legal code in Mongolian and printed it 1667. The following entry specifies the circumstances where the parent can rightfully arrange marriage as they wish:

[For the girl of] the negotiated betrothal, if [the man] does not take her after she reaches the age of twenty, let the girl’s father and mother be allowed to decide where to give her on their own will. ²⁰

According to the code of 1667, parents are allowed (or enjoy the right) to act according to their own will so long as they do not transgress the established norms of the community, as further confirmed by the promulgated laws of the state.

The negative valorization of liberty, which considered acting on one’s own will as arbitrary and licentious, formed another legacy of the global network of liberty of medieval times. Researchers have shown that in many parts of the medieval world, including Tang China, the Byzantine Empire, and post-Carolingian Europe, powerful aristocrats and/or officials defended their liberty to do “bad” things, such as pillaging or exploiting the people; against this backdrop, criticism of such abuse of “liberty” formed a critical part of the development of political thought in these regions (see Yin 2022, esp. 535, 553; see also Coulson 2003, 98; Koziol 2018, 77). Recently, a growing body of literature has shown that across the medieval and early modern globe, various mirrors for princes advised rulers to exercise caution against the temptation to act on their will (Forster and Yavari 2015; Blaydes, Grimmer, and McQueen 2018; Yin 2023). Nevertheless, the negative valorization of liberty has long been overlooked under the prevailing Euro-American tradition, which portrays liberty as an exclusively noble ideal. The network approach allows us to rethink the complicated relationship between the positive and negative values associated with the conceptual complex and further broaden the archival base of the history of political thought.

As shown above, bilingual or multilingual bureaucratic manuals, which have remained understudied by political theorists and historians, prove to be particularly promising for future research. To give yet another example, the bilingual Manchu-Chinese bureaucratic manuals promulgated in Qing China specifically castigated officials who would “sit quietly at their own will” (ini cihai ekisaka tehe/偃息自由, Liubu chengyu 1842, 1.16). These officials were not supposed to behave at their liberty even if they were not breaking the written codes of the empire; instead, they were expected to actively devote themselves to their duties and the common good. In this sense, taking actions (or taking rest) on one’s own will implies that the involved official is self-interested, precisely the connotations that the Persian term khud-murād carries. The implication here is that a polity needs to constrain those in power from such licentious acts of liberty so that the entire community can enjoy their liberty. In the late eighteenth century (ca. 1794), the Manchu court compiled the so-called Pentaglot Dictionary, which connects the Manchu notion of following one’s will (cihai) to the word of Chaghatay, the Eastern Turkic language—that is the predecessor of modern Uzbek and Uyghur: maylī میلی (Wuti qingwen jian 1794)—meaning inclination. How this notion is related to the Turkic words such as serbestiyyet (later also serbesti) “connoting the absence of limitations or restrictions” awaits future research (Lewis 1986, 589; see also Lewis 2008, 153n2). In any case, the resulting findings will not only deepen our understanding of the political vocabularies of the Turkic tradition but also invite attention to the connections between the Turkic and Manchu worlds.

Given the word limit, it is impossible to delve deeply into any concept of liberty or related words mentioned in this section. My goal is not to be comprehensive but rather to elaborate on a way of thinking about local changes and global transformation—that is, the relationship between nodes, clusters, and the totality of the network. Following the connections drawn by medieval translators, I show that it is productive to understand the myriad conceptions of liberty through the lens of positive versus negative valorization. With the former, liberty was considered as important to possess. Thus, the states—as evidenced by both England and China—promulgated codes to protect the liberty of their subjects to act freely within the scope of customary and written laws. With the latter, or negative valorization, liberty was considered dangerous to possess. Thinkers exemplified by Chen Gui under Jurchen rule or Shaikh Meḥmood in Delhi advised monarchs to be cautious of the liberty to act according to their will, a supremely privileged position they enjoyed. These two perspectives are not inherently contradictory. As illustrated above, for other members of the political community to enjoy liberty as a positive force, the liberty of those in power must be cautioned against and constrained. Political actors and thinkers of a particular time and space were addressing this shared continuum with increasingly connected terms. In this sense, liberty is as integral to the globally shared tradition of medieval connections as it is to well-known Western traditions, including liberalism and republicanism.

CONCLUSION

This article has elaborated on the network approach to political concepts that takes “connection” first. I have demonstrated its merits from three three perspectives. First, and foremost, it reveals that in the late medieval world, a global network of words and concepts of liberty had already emerged. Second, recognizing this network directs our attention to key nodes or clusters of terms—exemplified by the Mongolian phrase analyzed above—whose significance in the history of political theory has long faded into oblivion. Third, the global network of liberty enables political theorists to rethink how ideas from a specific context (such as the Mongol Empire) formed a part of the global transformation where different thinkers navigated through the positive and negative valorizations of liberty.

How does the global network of liberty revise our understanding of the history of this concept then? Or, to put it bluntly, if the Enlightenment concept of liberty restricts state power and protects individual rights, how might the medieval ideas and valorizations identified above contribute to a better form of political life? My response is twofold. First, I have proposed that a variety of concepts from the medieval network seek to restrict whoever is in power in their own ways, and the Enlightenment ideal of protecting individuals from “government encroachment” (de Dijn 2020, 4) is better understood as a way of carrying on the globally connected legacy. Second, I have suggested that the Enlightenment conception of liberty has unfortunately obscured other approaches to valorizing liberty. For instance, it is the global network of liberty that constantly reminds us that liberty can be a great thing to enjoy but also a dangerous privilege if abused by those in power.

Last but not least, although the post-medieval transformations exceed the scope of this article, it is important to emphasize that the situation changed dramatically after “the Europeans were finally able to join the era of globalization inaugurated by the Mongols in the thirteenth century” (Matthew Melvin-Koushki 2016, 148). With the global expansion of European and later American empires, as Wael Abu-‘Uksa (2016, 18) and other scholars (Immerman 2012; Pitts 2005) have demonstrated, writers of the British, French, and American empires appropriated the positive valorization as an exclusive Western heritage and used this to further their imperial ambitions of colonizing the rest of the world. During this process, thinkers in the colonies redeveloped the positive valorization of liberty to fight for individual freedom and collective independence (e.g., Sultan 2020). I envision that the network approach to concepts can offer a productive framework for understanding this transition, as well as the transformations of other political concepts. The systematic study of these connections has the potential to reveal that these ideas and practices in question are “a product of the world, not just of the West,” using Sarah Schneewind’s terms (2012, 91). On that day, we will likely be in a position to re-assess how the expansion of Euro-American empires, besides (or instead of) facilitating the globalization of liberty and other values, undermined the development of pre-existing conceptual networks.