How does philosophy learn to speak a new language?

ABSTRACT How does philosophy learn to speak a new language? That is, how does some particular language come to serve as the means for the expression of philosophical ideas? In this paper, I present an answer grounded in four historical case studies and suggest that this answer has broad implications for contemporary philosophy. I begin with Jonathan Rée’s account of philosophical translation into English in the sixteenth century, and the debate between philosopher-translators who wanted to acquire – wholesale or with modifications – foreign terms, and those who wished to take existing words and transform them from their ordinary to a philosophical use. I sketch how these twin processes of ‘acquisition’ and ‘transformation’ manifested themselves in philosophical translations from Greek to Latin, Greek to Arabic and both Greek and Arabic to Gə’əz and argue that comparative work in this vein could yield interesting and significant results. I suggest that not only is Rée’s approach useful for thinking about philosophical translation historically, but that philosophical translation between very different languages is important for contemporary philosophy insofar as it reveals the linguistic presuppositions of philosophical theories expressed in some particular language, and that this constitutes an argument against the prevailing monolingualism in philosophy.


Introduction
The notion of philosophy 'speaking' is perhaps less strange an idea than it initially seems.Philosophia was often personified in antiquity as a woman, and the idea of teaching her to speak languages other than her native tongue of Greek was introduced by one of the earliest and greatest of philosopher-translators, Cicero, who characterised his project as 'teaching philosophy to speak Latin', 1 not only by importing originally Greek words (dialectica, physica, politica) into Latin, but by teaching philosophy new terms (moralia, naturalis, ratio) from his native language.Fifteen centuries later, the preface to the first translation of an entire work of philosophy into English -fittingly enough a translation of Ciceroproclaimed that just as Cicero had 'clad ladie Philosophie in Roman attire', the new translation had 'caused Marcus Tullius [Cicero] (more than he could do, when he was alive) to speake English' (Quoted in Rée, 2018, p. 30).
How does philosophy learn to speak a new language?That is, how does some particular language come to serve as the means for the expression of philosophical ideas?In particular, how is this possible, where it requires the forging of concepts in a language that is formerly without them?This paper presents an analysis of historical instances of philosophical translationof philosophy coming to speak a new languageand proposes a project of extending such an analysis to as many languages and traditions as possible.We begin with the example of philosophical translation in sixteenth-century England as outlined in the opening chapter of Jonathan Rée's Witcraft: The Invention of Philosophy in English.I suggest that Rée points towards a fruitful way of thinking about historical cases of philosophical translation, in terms of a debate between proponents of 'acquisition' and 'transformation', and that we might think of philosophy learning to speak a new language in terms of the construction of a philosophical-conceptual vocabulary by means of the acquisition of foreign terms and the transformation of existing terms to new meanings.
I then suggest that this framework would be helpful for understanding how philosophy came to speak other languages, and thus for providing an answer to our question more generally.I sketch how the twin processes of 'acquisition' and 'transformation' are at work in philosophical translations from Greek to Latin, Greek to Arabic and both Greek and Arabic to Gə'əz and argue that comparative work in this vein could yield interesting and significant results.I will suggest that not only is this distinction useful for thinking about philosophical translation historically, but that philosophical translation between very different languages is important for contemporary philosophy insofar as it reveals significant linguistic presuppositions of philosophical theories expressed in a particular language, and that this constitutes a strong argument against the tendency towards monolingualism of a good deal of contemporary philosophy. 2 1.1.'We English men have wits' Jonathan Rée's Witcraft: The Invention of Philosophy in English traces an unorthodox, international history of philosophy in the English language from its origins in Elizabethan England to Wittgenstein, Ramsey and Russell in the mid-twentieth century.Its opening chapter takes as its focus the birth of philosophy in the English language, examining how the Europe-wide processes of the translation of theological, literary and eventually philosophical texts from cosmopolitan Latin into local vernacular was accomplished in the case of England and the English language.The chapter begins with mediaeval arts course Aristotelianism and its scholastic Latin philosophical vocabulary of 'quiddities', 'haeccities' and the like, and with the revolt of the classically inspired humanistsalso writing in Latinagainst this high scholastic style.Rée then traces the pre-history of philosophical translation in English in the form of translation of scripture and the subsequent explosion in the profusion of philosophical texts in English in the sixteenth century.
Debates over the correct translation of some philosophical term of art were foreshadowed by arguments between translators of scripture: did the English 'church' mean the same as the Latin congregatio?Was it legitimate to translate the originally Greek baptismata as 'washings'?(See Rée, 2018, p. 19).How far was it legitimate to depart from original terminology in the service of increasing the understanding of a text?Practically speaking, once translators had experience in working on the transmission of technical terms from Latin and Greek, two languages that were the vehicles not only of scripture but of philosophy, it was but a short step to begin translating works of philosophy into the English vernacular too.
These translations began in 1547, when William Baldwyn published a Treatise of Morall Philosophye, contayning the Sayinges of the wyse, a compendium of sayings from ancient philosophers, based on the Greek text of the Lives and Opinions of Eminent Philosophers by Diogenes Laërtius.The first translation of a philosophical classic in its entirety (as opposed to a selection of maxims) was the aforementioned Ciceroes three bookes of duties (De Officiis) published in 1558.This wasquite unusuallypreceded by the first original work of philosophy, Thomas Wilson's 1551 logic textbook The Rule of Reason, which aimed to bring logic into English for the first time.Wilson began by providing definitions of foreign terms in English, such as logique ('an Arte to try the corne from the chaffe, the truthe from every falshed'), Genus ('the general worde'), Species ('the kinde'), Accidens ('the thing chauncing or cleving to the substance'), as well as some rather more metaphorical suggestions, including that of a dilemma as a 'horned' argument (Quoted in Rée, 2018, p. 19).
However, it is with two works published later in the century: Ralph Lever's Witcraft (from which Rée takes the title of his history) and Thomas Blundeville's Art of Logike that will be our focus in examining the complexities and challenges of philosophical translation, and that will serve as a model for the rest of the study.

'Every simple shewsay … is either a yeasay or a naysay'
In attempting to render Latinate philosophical terms in a Germanic language, philosopher-translators like Baldwyn and Lever were faced with a problem.Should they use 'terms of some other tongue … and by a little chaunge of pronouncing, to seeke to make them Englishe worded'that is, whether to acquire, wholesale or with modifications, these foreign words already laden with philosophical significance, or to coin 'understandable termes, compounded of true and auncient English words', that is, whether to take existing words and somehow transform them from their ordinary to a philosophical use (Quoted in Rée, 2018, p. 21).To acquire terms, or to transform them?
Lever was firmly on the latter side of the dilemma, and in Witcraft his transformations consisted in the coining of new compound terms from traditional Anglo-Saxon words and roots: Conclusiones became 'endsays', affirmationes were 'yeasays' and propositiones conditionales were 'ifsayes'.Instead of the quasi-Latin maxim 'every proposition is either an affirmation or a negation,' we could now say that 'every simple shewsay … is either a yeasay or a naysay' (Rée, 2018, p. 21).Lever justified these ingenious transformations of existing language on the ground of good linguistic usage: We … that devise understandable termes, compounded of true and auncient English words, do rather maintain and continue the antiquity of our mother tongue: than they, that with inkhorne termes doe change and corrupt the same, making a mingle mangle of their native speache, and not observing the properties thereof.(Rée, 2018, p. 21) Needless to say, a plain-speaking English was preferable to a supposedly obfuscatory and antiquated Latin in Elizabethan England for reasons that were not solely theoretical.Latin was not only the language of the old Aristotelian scholarship but of the old Catholic church (even if Protestant thinkers continued using Latin for a good whilethe Pietist Kant's Inaugural Dissertation of 1770 was composed in Latin), and authentic, Germanic terms could easily be associated with patriotism and Protestantism. 3 One side, after all, 'maintain and continue … our mother tongue', whilst the other 'change and corrupt the same'.Robert Cawdrey, in his A Table Alphabeticall goes as far as to compare this form of linguistic mixing to 'counterfeiting the King's English'. 4And if even scripture could now be translated into vernacular English, why not philosophy too?
Blundeville on the other hand did not propose to 'fayne new words … as some of late have done', considering it 'no shame nor robberie to borrowe tearmes … from the Latines, as well as they did from the Greekes'.If borrowing words was good enough for Cicero, it was good enough for the English philosopher.There was no need for Lever's clumsy coinageswe should acquire terms from Latin and Greek directly. 5 Many of these terms would over the coming decades and centuries percolate from logic textbooks and philosophical treatises to everyday language.Evidently the foreign imports had prevailed by the time John Florio, the translator of Montaigne claimed that he could not 'philosophate' and 'fantastiquize' in English without borrowing 'uncouth words' from French and Latin, and the disuse of Lever's home-made 'yeasay' and 'naysay' (for affirmatio and negatio) compared to the overwhelming everydayness of terms owed to philosophical translation like 'nature', 'ethics' and 'politics' is the strongest evidence for this.

Acquisition and transformation
This debate between those who would coin new terms and those who would import foreign terms illustrates two ways in which philosophical translation teaches philosophy to speak a new language: acquisition and transformation.In 'acquisition' a language imports words that express philosophical concepts, that is, when it directly transcribes a term from another language, importing it into its own.As we have already seen, ordinary English is full of such terms, as are many of the vernacular languages of Europe.Perhaps the paradigm example of acquisition is the term for philosophy itself: almost all languages with a word for philosophia take over some more or less direct version leading back to the Greek original. 6For example, the English term 'philosophy' arrives from Greek by way of Latin, and the Gə'əz fälsafa arrives from the same source by way of Arabic.
The second way to express some new concept is to take a word and transform it from an 'ordinary' to a philosophical use by means of some essentially metaphorical tweaking of its meaning.Consider Lever's ingenious interpretation of 'dilemma' as 'a horned argument'.Even words which seem to have taken on a quite distinct technical meaning of their own are often revealed to have roots in a metaphorical transformation: take the example of 'substance', meaning whatever 'stuff' it is that underlies qualitative and quantitative change.Etymologically substantia means precisely 'something that stands under or grounds things', but few philosophers would argue that their candidate for substance is the sort of thing that could really 'stand' in any literal sense, and certainly not 'under' or on the 'ground'.It is a concrete term that has been stretched beyond its standard usage to express a concept that cannot be expressed with the existing set of words in their ordinary usage.
Other more recent examples of metaphorical transformation might be the term 'ground' in contemporary analytic metaphysics, the notion of 'reduction' in the philosophy of science or 'construction'as in 'the social construction of x'in various contexts.But some transformations are not intentional metaphorical or technical adaptations of a word for a particular purpose so much as accumulations of many years of using words in different ways, as a result of the unusual contexts familiar words find themselves in during the process of philosophising.Consider the fact that familiar termstruth, being, mindcome to be used in a wide variety of quite technical and unusual ways in philosophical literature.We might think about this as the difference between the ordinary use of 'truth' as in 'tell the truth' or 'the truth hurts', and the technical theory of truth as correspondence or coherence or whatever else, what we might call philosophical truth, or 'Truth' with a capital T.
Transformation and acquisition are of course heuristics that in their simplicity conceal a wide range of translational practices.Acquisition may take the form of a transliteration, a semantic or partial calque or a more direct borrowing.Transformation on the other hand may involve the metaphorical tweaking mentioned above; using a familiar word in an obscure, arcane or archaic way; combining familiar terms in the style of Ralph Lever; providing an arbitrary new definition for the term (consider the 'grue' and 'bleen' of Nelson Goodman (Goodman, 1955), the 'qaddition' of Kripke's rule-following paradox) (Kripke, 1982).It is not always clear which of these two options the translation of some given term is, and these are certainly not the only means of translating terms, but these two heuristics are suitably general as to cover some of the variety of practises that occur in the different contexts we will examine and thus provide the basis to engage in comparative work on very different languages.
Although terms appear in a new language by means of either acquisition or transformation, once they have reached the language, there is no hard and fast distinction between the two.Acquisitions may in turn be transformed: both psyche and logos are Greek terms, but the Greek did not have a word for a separate branch of knowledge of 'psychology'.Terms transformed in one language can be acquired by another: the English 'substance' is an acquisition from the Latin transformation of substantia, possibly via French.These two techniques of acquisition and transformation furnish a set of philosophical terms in a language that did not have them before.We might term this setthe set of philosophical terms of art not found in ordinary language, or extended beyond their ordinary usagea philosophical-conceptual vocabulary.

Polyglot philosophy
How does this work in other examples of philosophical translation?That is, how does acquisition and transformation allow for the construction of a philosophical-conceptual vocabulary?In the interest of demonstrating the usefulness of this framework I will select three examples that are distant from our initial case in either time, space, or linguistic structure: the translation of Greek philosophy into Latin in the first century BCE, the translation of Greek philosophy into Arabic from the eighth century, and the translation of Greek philosophy from Arabic into Gə'əz in the sixteenth century.We begin this section as we began the paper, with Cicero.

'Ladie Philosophie in Roman attire'
Cicero aimed to teach philosophy to speak Latin, and in doing so to assert Latin as a philosophical language, challenging the philosophical hegemony of Greek.As Carlos Lévy notes, this was a radical gesture, going against the prevailing wisdom of the international world of Hellenistic philosophy, composed in Greek from Rome to Alexandria and Ai Khanoum.Cicero's philosophical translation, like Lever's, had a political edge.Horace famously declared that 'captured Greece has conquered its savage captor' (Graecia capta ferum victorem cepit et artes intulit agresti Latio), and Cicero clearly felt that in teaching philosophy to speak Latin, in demonstrating the aptitude of the Latin language for original philosophising, he was continuing the work of the legionaries, subduing 'the last cultural province that still escaped Roman hegemony' (Lévy, 2021, p. 75).
In doing so, Cicero displayed a marked preference for transformation over acquisition.Although a good many terms are imported directly from Greek by transliteration (rhetorica, dialectica, geometria, or musica and, of course, philosophia), most of these were already in circulation by the time of his earliest translations, and 'we may consider as being our own; the ideas might it is true have been translated into Latin, but the Greek terms have been familiarized by use' (Cicero, 1914, p. 221).This approach was certainly suggested by the patriotic politics of his project, but also asserted that his own language was full of semantic possibilities, was perfectly capable of expressing any idea it required, given the efforts of a suitably talented philosopher-translator: 'in fullness of vocabulary we are not merely not surpassed by the Greeks but are actually their superiors' (Cicero, 1914, p. 221).
Translation, for Cicero 'was also a means to give a post-mortem life to ancient practices' as evidenced by his translation of oikeiosis (οἰκείωσις), the central term of Stoic ethics (Lévy, 2021, p. 83).While it would have been perfectly possible to translate this notion with a neologism like domesticatio (a calque of oikeiosis), he opted for conciliatio and commendatio, terms which formed a part of the political rhetoric of his day and thus held living resonances that already-ancient Greek terms could not have for a Roman audience.
Another major difficulty in the translation of Greek ethics was the term 'kalon' (καλόν), a word with a semantic field encompassing 'good', 'fine', 'beautiful', but used especially in philosophy for expressing ethical perfection.In a Hellenistic context, the notion had a strong aesthetic connotation, such that 'beauty was the external face of a virtuous interiority' (Lévy, 2021, p. 82), but in Cicero's rendering of the term by the Latin word honestum he chose to express the ethical ideal not in the language of aesthetic beauty, but interpersonal, social ethics.
In epistemology, Cicero was troubled by how to translate the term epochē (ἐποχή), 7 the 'suspension of judgment' or 'withholding of assent' introduced by Pyhrro and central to the Academic Scepticism favoured by Cicero.In a letter to his friend Atticus (Att.13.21), Cicero wonders about the best translation of the term.There is the verb dubitare ('to doubt'), as used by Lucretius in De rerum natura, but the Academic notion of epochē suggests a rather different conception of mental functions from Lucretius' Epicurean system.Then there was Atticus' proposed choice of inhibere ('to prevent'), which Cicero originally went along with, but on consulting a group of local sailors who informed him that inhibere referred in their profession not to the blocking of oars in the water, but to an alternative method of rowing, he changed to yet another alternative verb sustinere ('to hold back') (Lévy, 2021, p. 76).
Cicero both painstakingly transformed everyday Latin terms for new, technical meanings in order to convey the complexities of Greek language philosophy, and (reluctantly) imported Greek terms or used Greek loanwords already in circulation.As we will see this process was not unique to English or to Latin philosophical translation but spanned many other language families and literary traditions.

Philosophia and falsafa
Moving closer in time, but further in space from our original case, we turn to one of the best-known periods in the history of philosophical translation: the Arabic translation movement.From the eight century Abbasid elites engaged in a massive project of translating philosophical materials, a project which was so successful that by the tenth century an educated resident of Baghdad could read almost all surviving works of Aristotle in Arabic.
Acquisition and transformation were once again at work in the process of creating a philosophical-conceptual vocabulary in Arabic.Though few acquired words penetrated into general usage in the way they did in English (though one example would be hāyūlā ( ‫ه‬ ‫ي‬ ‫و‬ ‫ل‬ ‫ى‬ ) from Greek hyle, which remains a term for 'matter', in the sense of 'stuff'), many philosopher-translators would transcribe or calque significant technical terms, sometimes providing an Arabic near-equivalent to make the meaning clear. 8 One example is, again, the term for 'philosophy' itself, taken over in a very direct way as falsafa ( ‫ف‬ ‫ل‬ ‫س‬ ‫ف‬ ‫ة‬ ).However, this doesn't come to describe the activity in the same way as in Latin and languages like English which are downstream of Latin in their translation history.Here the term falsafa is used to distinguish the tradition of thinking with and after the Greeks from ʿIlm al-kalām ( , lit.'science of discourse'), the tradition of Islamic rational theology, a discipline very much like philosophia in many of its topics (the soul, causation, free will, moral responsibility, etc.) and forms of argumentation, but distinct in its historical source.Peter Adamson has argued that 'if there had been no translation movement, historians would just think of kalām as the philosophical tradition that can be found in medieval Islam' (Adamson, 2019).
To further complicate the picture, another term, hikma ( ‫ح‬ ‫ك‬ ‫م‬ ‫ة‬ ), meaning 'wisdom' functioned in the same semantic realm, and the agent noun hakīm, translatable as 'wise man', 'sage' or 'philosopher', could easily be applied to a figure like Aristotle or Socrates.Like in English, but apparently unlike in Latin, the calque of philosophia never lost its foreign flavour, and this leant a political shade to the translation.Philosophia remained a Greek, Christian, idea, and therefore 'some authors would self-identify as doing hikma but not falsafa, because they are thinking about philosophical and theological issues in a rational way but not engaging directly with Greek sources in Arabic translation' (Adamson, 2019) Perhaps for this reason, Arabic and Persian terms tend to crowd out Greek imports over time.

PERSPECTIVES
A more significant difficulty for the transfer of philosophy into Arabic resulted from the dramatically different grammatical structures of Arabic and Greek.Arabic, as a Semitic language of the larger Afro-Asiatic family is not at all related to Greek, a language isolate within the Indo-European family.Many of the translators into Arabic could rely on earlier translations from Syriac (also a Semitic language) who had in fact done the initial groundwork, the linguistic 'heavy lifting' as it were, in grappling with how to translate from Greek into a Semitic languages, but problems nevertheless remained.
Perhaps the most significant of these was what to do with the grammar of the Greek verb einai (ειναι), 'to be'.There is no equivalent to the various uses of einai and its cognates on (ον) and ousia (οὐσία), in Arabic or other Semitic languages, with their application being very general and diversified compared to the verb 'be' in other language families.This will be especially important to examine in light of my later suggestions about the significance of translation across very different languages, and because such questions are largely beyond scope of Rée's analysis, focusing as it does on Indo-European languages.We will return to the significance of this issue in section 3.1.

Philosophy learns Gə'əz
Gə'əz 9 is an Ethiosemitic language spoken in northern Ethiopia from antiquity until approximately the thirteenth century.Philosophy in some form or other is attested in Gə'əz literature from as early as the fifth century in translation, 10 with a sharp increase in manuscripts from the fifteenth.Philosophy learned to speak Gə'əz at almost exactly the same time it was learning English, during the second great flourishing of Gə'əz literature in the sixteenth century.
The main source of philosophy in this period is the Book of the Wise Philosophers, a compendium of philosophical maxims very much like Baldwyn's Treatise of Morall Philosophye, originally a translation of a Christian Arabic gnomologium, now lost, based on the Nawādir al-falāsifa (Apophthegms of the Philosophers), of the ninth century Nestorian Christian H  unayn ibn Ish  āq.Unusually for a Gə'əz text, the identity of the translator and date of the translation is known.Between 1510 and 1522 one abba Mika'el 'the translator', a priest of Egyptian origin translated the text 'by mouth', that is, by orally translating the Arabic text into Gə'əz, with this text being copied by an Ethiopian scribe.
Just as with English, the translation of scripture served as the pre-history to the translation of philosophy in Gə'əz. 11Both equipped generations of translators with the knowledge of multiple languages that were vehicles not only of scripture but of philosophy, and once scripture was translated, there was no reason that the rest of literature, including philosophy, could not follow.Early examples of both acquisition and transformation can be found in this translation tradition.At the level of a direct translation, many Greek terms are transcribed directly from the Greek or the Latin, and the significant number of Greek imports are visible from the addition of the two new letters to the Gə'əz writing system -'ፐ' and 'ጰ'specifically for the purposes of rendering the 'p' sound used in a number of Greek words.A more sophisticated example of linguistic acquisition might be the calque of the Greek Theotokos (Θεοτόκος), meaning the 'bearer of God', as wäladitä 'ämlak (ወላዲተ አምላክ)this is a point where acquisition begins to meet transformation: the term has been rendered faithfully enough (a literal translation might be 'the mother of the lord'), given the difficulties in traversing the linguistic gulf, but the translator evidently felt that a calque that appeared somewhat familiar to its Ethiopian readership was preferable to a transliteration.'Philosophy' on the other hand, is once again itself an example of an acquisition, its translation in Gə'əz as fälasfa (ፈላስፋ) coming directly from the Arabic falsafa.
Transformation too is an essential component of theological translation in Gə'əz.The notion of 'humanity' or 'mankind', essential for the universalist message of Christian ethics, is translated into Gə'əz as 'əgwalä 'əm-həyaw (እጓለ፡እመሕያው) 'the offspring of life', suggesting the idea, possibly a remnant from pre-Christian Aksumite religion, of the origin of all peoples from a primordial mother-goddess.The line between theological and philosophical translation is blurred.On a much more abstract and metaphorical level, the Neoplatonic-derived notion of hypostasis (ὑπόστασις), often translated in philosophical contexts as 'substance' or 'essence' or in theological contexts as 'person' (as in the three persons of the Trinity) 12 is translated into Gə'əz as bähri (በሕሪ), lit.'pearl', suggesting the 'essence', 'nature' or 'character' of a thing that underlies change.Once this meaning was establishedonce the term was transformed from its everyday to its philosophical useit was available as part of the conceptual vocabulary of Gə'əz and could be taken up for use in the translation of philosophical treatises, even where the original Greek or Arabic term might be quite different from hypostasis.
Although these works are relatively well known by scholars of Gə'əz literature, they were not widely considered to form the basis of a philosophical tradition in Gə'əz until the pioneering work of Claude Sumner, who argued in the first of his fivevolume work Ethiopian Philosophy, that these texts not only formed the first works of philosophy in Gə'əz, but the first examples of active philosophical work in Gə'əz.
The distinction consists Sumner's claim that the Ethiopian Book of the Wise Philosophers is not merely the faithful transmission of foreign texts and ideas into Gə'əz, but that the works are transformed in the process, acquiring a distinctively Ethiopian character: 'this work is Ethiopian, not by the originality of its invention, but by the originality of its style and presentation' (Sumner, 1986, p. 29) This is, according to Sumner, because 'Ethiopians never translate literally: they adapt, modify, add, subtract.A translation therefore bears a typically Ethiopian stamp: although the nucleus of what is translated is foreign to Ethiopia, the way it is assimilated into an indigenous reality is typically Ethiopian' (Sumner, 1986, p. 29).
This, it seems to me, would have appealed to Cicero.His translation of Greek ethics consisted in adapting, modifying, adding and subtracting from the Greek original, shifting linguistic metaphors for ethical perfection from the aesthetic to the social, thereby printing his subject with a typically Roman stamp.The case is more difficult to make for logical theory, but might Ralph Lever too have approved of the idea that in transforming 'inkehorne terms' into 'understandable termes, compounded of true and auncient English words', he was impressing upon philosophical logic something of a distinctively English character?If the processes of acquisition and transformation changes not only the external linguistic form of the texts but something of their internal message, then philosophy learns to speak a new language by translating, and this translation is always itself a way of philosophising.
Sumner's suggestion shows one way in which translation constitutes active philosophical work, 13 and part of what I have been trying to demonstrate is the form this philosophical work took, namely a process of transforming ordinary terms from the native tongue into vehicles of philosophising, and the careful evaluation of when it was permissible to import a term wholesale from another language.This involved difficult philosophical and literary choices: what use of a word would best convey the sense of some foreign term?What kind of imagery might help to convey the meaning of this term in a quite different language?What were the many possible confusions?Did the translation of some term require one concept, or two?Or many?Ordinary language, no matter which particular natural language, does not typically have the conceptual resources to philosophise with, where philosophy is understood as abstract, more or less systematic theorising about fundamental matters.For any language to become a medium of 'doing philosophy', terms must be borrowed, invented, or wrenched from their everyday use.In Cicero's words, 'philosophy is the Science of Life, and cannot treat its subject in language taken from the street' (Cicero, 1914, p. 221).
This work demanded both linguistic and philosophical skill, and also the navigation of the cultural politics of translation: too many Greek terms in his translations, and Cicero would have conceded Greece victory in the cultural struggle; in sixteenth-century England, too many Latinate terms were liable 'change and corrupt' whereas Germanic ones 'maintain and continue'; translators in the early Islamic caliphate had reason to consider the opinion of the religious establishment as they navigated the choice of translation between hikma, falsafa and kalām.The significance of cultural politics to philosophical translation was an unexpected result of this enquiry and deserves significant future study in its own right.

The significance of philosophical translation
Why study philosophical translation?First there is the obvious benefit to the speakers of the target language.The cases we examined enabled millions of English, Arabic, Gə'əz and Latin and speakers to engage in philosophical reflection in their own language, and spawned immensely fruitful, centuries-long traditions of philosophical writing in those languages.In teaching philosophy to speak these languages, philosopher-translators made philosophy accessible to their speakers, and in this sense, translation is an egalitarian gesture, even when inspired by nationalistic considerations.
There are also benefits for speakers of the source language.Consider Cicero's translation of kalon as honestum.This not only allowed Latin speakers to get a grasp on Greek ethics but revealed to the Greeks an alternative way of understanding kalon as ethical ideal, in terms of interpersonal responsibility rather than aesthetic excellence.In expanding the range of possible interpretations and meanings of the term, translation draws attention to the cultural-linguistic specificity of the aesthetic connotation of kalon, thereby enriching the original concept.
To examine what I take to be a major benefit of philosophical translation to philosophy more generally, we need to return to the challenges faced by Syriac/Arabic translators in crossing the boundary between Indo-European and Afro-Asiatic grammars.In Section 2.2 we noted the difficulty of translating the extremely varied uses of einai and its cognates on and ousia into Semitic languages, in which the verb 'be' functions very differently, with a much more limited range of applications.
The French linguist Émile Benveniste famously suggested that 'the systematic development of a concept of Being in Greek philosophy […] relies upon the pre-existing disposition of the language to make a very general and diversified use of the verb einai', 14 that is, that the grammar of the Greek language predisposes (but does not determine), particular conceptions of ontology.Benveniste continues, 'it is generally recognized that this wide range of uses for the single verb einai in Greek reflects a state of affairs which is peculiar to Indo-European languages, and by no means a universal situation or a necessary condition.'(Benveniste, 1966, p. 73).The grammar of the Greek verb 'to be' suggests particular conceptions of ontology, conceptions that might not seem natural or indeed even correct to speakers of languages with quite different grammatical predispositions. 15 That there may be quite significant alternatives to Greek grammars of 'to be' and thus to traditional ontological categories, is suggested by a number of linguistic as well as philosophical sources (See, for example, Basson & O'Connor, 1947).Charles H. Kahn in his classic The Verb 'be' in Ancient Greek, wrote that 'the present monograph on "the verb 'be' and its synonyms" shows just how far the languages of the earth may differ from one another in their expression for existence' (Kahn, 1973, p. 1).If this is the case, might the grammatical structures of Afro-Asiatic, Dravidian, Sino-Tibetan or Uralic languages suggest alternative parameters for the development of a concept of being?And what better way to explore the nature and extent of this purported different than by examining how such a concept translates across linguistic borders?
For example, in examining how an ontological theory grounded in a particular set of predispositions comes to appear in a language that does not share them, we might examine how mediaeval Aristotelian philosophers who did not use Indo-European languages couched their own ontological speculation, and how they translated Aristotle's.We might examine how the interpretations of Avicenna and Maimonides differed from those of their counterparts in the Latin world, who examined the very same texts in translation, but with the difference of reading it in another Indo-European as opposed to a Semitic language.
Why should this be of anything other than historical interest?For philosophers who consider the deliverances of ontological speculation to be general, unchanging and universal, this difficulty appears as a kind of sceptical challenge: a theory of categories, or of primary substance, is supposed to apply universally, not only to the speakers of some languages and not others.But if the concept of 'being' in Western philosophy relies on pre-existing dispositions of a particular language or set of languages, and if those dispositions are not universal, why should we suppose that the theories constructed using the conceptual vocabulary based on those dispositions are universally valid?Another way of putting this is to ask whether the deliverances of philosophical ideas expressed in some conceptual vocabulary can possibly be as universally valid as they purport to be if they depend crucially on artefacts of sedimented linguistic usage.How are we to know that the concepts that philosophers believe 'carve nature at the joints', are not simply 'innocent linguistic categories […] taken on the formidable appearance of cosmic absolutes' (Sapir, 1924, p. 154).
To be clear, this is not intended as a defence of linguistic determinism. 16Far from it.The idea is not that a language or language family determines a philosophical ontology, far less a comprehensive 'worldview'.To adapt a famous argument from Davidson (1973), the manifest success of philosophical translation should serve to refute any such determinism.The suggestion is rather that in order to better understand our own conceptual vocabularies, and how far they apply, we would do well to consider where our words that express these concepts come from, their histories, and how people elsewhere and at different times might have tried to express them.
A sustained comparative examination of philosophical translation, both within and across different language families, is therefore of significance not merely for the history of philosophy and for translation studies, but for assessing the purported universality of the very tools of our philosophising today.If philosophy is to be truly presuppositionless, more needs to be done to examine the way in which our philosophising in some particular language is led in certain directions by particular local factors of that language.The only way we can do this is by comparing this language to others, and the best way to accomplish this comparison is by translating, and translating across as wide a range of languages as possible.
We have briefly examined four examples from three continents and as many millennia, but the history of philosophy is littered with examples of a far greater range.Some, like the translation of Greek philosophy into Latin and Arabic or from Latin into French, English and German, are well-known and well-studied.Others, like the translation of philosophy from Greek into Syriac, Armenian, Georgian, from Arabic into the Ajami scripts of West Africa, from Sanskrit and Pali into Tibetan, Chinese and the Indian vernaculars are less well-known.
We are evidently (and unfortunately) moving the opposite direction.An increasing majority of scholarship is composed in English, a language that, as we have seen has a capacious and diverse conceptual vocabulary, but which is a single language nevertheless.Not only does working with a single language deprive us of a major source of the lifeblood of new words and new concepts, but if my above argument about the significance of translation for assessing the linguistic presuppositions of our concepts is correct, this monolingualism can only result in a myopia of unexamined presuppositions.
In each of our case studies, we saw a symbiotic relationship between philosophical translation and the target language: philosophical translation enriched English, Latin, Arabic and Gə'əz, and the conceptual resources of these languages in turn enriched philosophy.I think that this constitutes a strong argument for the importance of linguistic diversity in philosophy, and a further argument for the importance of philosophical translation now.In addition to the critical task of revealing the linguistic specificities of particular concepts, philosophical translation creates new concepts and conceptual vocabularies.
Indeed, a broad enough study of the interrelations of the conceptual vocabularies of very different languages might serve as a way of achieving another kind of universality: not the overarching universal of a one philosophical-conceptual vocabulary defeating all others, but what Souleymane Bachir Diagne has called the 'lateral universal' of ethnographic experience, the incessant testing of the other through the self, and the self through the other. 17This would be a universal borne not of the victory of a single hegemonic vocabulary, but an ideal of universality as the set of possible translations between a manifold of vocabularies.
2. In speaking of a 'tendency towards monolingualism', I have in mind the increasing tendency for philosophers who might have written in their own language even 30 years ago to feel obliged to write in English.As English continues to grow and a language of global scholarship, and as analytic philosophy continues to gain prominence in non-Anglophone countries, more and more philosophy ends up being done in English.As Schwitzgebel et al. (2018) have shown, 'in a sample of papers published in elite journals, 97% of citations were to work originally written in English.73% of the papers in the sample didn't cite any paper that had been originally written in a language other than English, and 96% of the members of elite journals' editorial boards are primarily affiliated with an Anglophone university'.For the original data, general discussion of the increasing monolingualism of philosophy, see the various papers from Philosophical Papers, Volume 47, Issue 1 (2018).Even an unusually multilingual and linguistically attuned work such as Cassin's (2004) Dictionary describes English as the 'dominant philosophical language', p. 257. 3. Indeed, notions of the contrast between the supposed opacity of 'continental' philosophical prose as opposed to the transparency of English language philosophical writing continued well into the twentieth century and lie at the root of many characterisations of the distinction between continental and analytic philosophy couched in terms of clarity or obscurity of prose.4. Cawdrey (1617).Indeed the concern about the import of foreign terms seems to have arisen in a context of heightened anxiety about the import of foreign currency, and would make for an interesting comparative study.5. Similar debates took place in other European countries around the same time.In Joachim Du Bellay's Défense et illustration de la langue française (1549), he sketches techniques for 'enriching' the French language, and explicitly recommends adopting foreign terms as one important way of doing so.6.According to Crane (2015), it is only Hungarian (bölcselet) and Dutch (wijsbegeerte) which have coined their own termsboth of which are calques of the etymology of philosophia.7. Interestingly, when Husserl resurrected the term in the early twentieth century, he left the term untranslated, simply glossing it with the German term Einklammerung, or 'bracketing'.8. Arabic is transliterated according to the IJMES transliteration system for Arabic, Persian, and Turkish.9. I employ the field-standard method of transcription of Gǝ‛ǝz terms outlined in the Encyclopaedia Aethiopica.Quotations always retain original method of transcription.10.For the earliest translations, see Sumner (1974).11.Though this pre-history is a lot deeper in the case of Ge'ez, with the first translations of scripture taking place almost a millennium before the translation of philosophy.12.This notion is crucial to Christian theology of the Trinity and especially for the Miaphysitic Christology of the Ethiopian Orthodox Church, which holds a non-Chalcedonian conception of the nature of Christ, professing that the Human and Divine natures (physis) of Christ are married into one hypostasis 'without separation, without confusion, and without alteration ' Farrington (2006).13.Cicero evidently agreed with this position, arguing in the Tusculan Disputations XVIII that he is no mere translator, serving as one only temporarily: 'I will perform on this occasion the office of a translator'.14.Kahn paraphrasing Benveniste in Kahn (1973, p. 1).15. Considering that he is such a significant figure for the myriad contributors to the Dictionary of Untranslatables, this cross-culturally comparative direction of extending Benveniste's work seems to be thus far quite underexplored.One hopes that the proposed extension of the Dictionary project to non-European languages will provide opportunity to further explore these possibilities.16.Indeed, it seems to me that the indubitable reality of translation (as opposed to its mere possibility) is the strongest argument against such determinism.17.Diagne (2013, p. 18).See also his argument that 'paraphrasing Umberto Eco who famously said that the language of Europe is translation, it could be said more globally that the language of the universal is translation.To acknowledge that is to abandon the assumption that the exploration of a supposed universal grammar of the Logos needs to be conducted in the silence of the empirical languages that humans actually speak, while some of them, the European ones, especially ancient Greek and German according to Heidegger, can appear as its realization, in some respect ' Diagne (2013, p. 15).

Disclosure statement
No potential conflict of interest was reported by the author(s).