1.Introduction

The nineteenth century was a period of dramatic change in Europe for the idea of history (Collingwood 1946Collingwood, Robin 1946 The Idea of History. Oxford: Oxford University Press., 126). From antiquity through to the eighteenth century, historiography had broadly been considered an artistic and rhetorical activity, and histories themselves were generally grouped together as a particular genre of literature (Morley 2014Morley, Neville 2014 Thucydides and the Idea of History. London: I. B. Tauris. , 20; Muhlack 2011Muhlack, Ulrich 2011 “Herodotus and Thucydides in the View of Nineteenth-Century German Historians.” Translated by Neville Morley. In The Western Time of Ancient History: Historiographical Encounters with the Greek and Roman Pasts, edited by Alexandra Lianeri, 179–209. Cambridge: Cambridge University Press. , 182). Texts such as Plutarch’s Lives or Xenophon’s Cyropaedia were thus read not so much as a means of gaining precise knowledge of earlier times, but primarily as sources of political and moral instruction, providing guidance for the present through past example (Pade 2006Pade, Marianne 2006 “Thucydides’ Renaissance Readers.” In Brill’s Companion to Thucydides, edited by Antonis Tsakmakis and Antonios Rengakos, 779–810. Leiden: Brill. , 779; Shapiro 2000Shapiro, Barbara 2000 A Culture of Fact: England, 1550–1720. Ithaca, NY: Cornell University Press., 35). Moreover, while it was generally expected among pre- and early modern readers that the historian ought not to write falsehoods, the literary-artistic qualities of a history ranked above the veracity of the account of the past it communicated (Muhlack 2011Muhlack, Ulrich 2011 “Herodotus and Thucydides in the View of Nineteenth-Century German Historians.” Translated by Neville Morley. In The Western Time of Ancient History: Historiographical Encounters with the Greek and Roman Pasts, edited by Alexandra Lianeri, 179–209. Cambridge: Cambridge University Press. , 186). Indeed, Muhlack (2011Muhlack, Ulrich 2011 “Herodotus and Thucydides in the View of Nineteenth-Century German Historians.” Translated by Neville Morley. In The Western Time of Ancient History: Historiographical Encounters with the Greek and Roman Pasts, edited by Alexandra Lianeri, 179–209. Cambridge: Cambridge University Press. , 182) has argued that historical truth was not considered to exist in and through itself before the modern period, but only in and through the internal coherence of its narrative expression: “truth was established in the literary work of art that the historian produced, fixed on the rules which he followed therein.”

In the nineteenth century, and largely as a result of the pioneering interventions of the German historians Barthold Georg Niebuhr, Leopold von Ranke and Wilhelm Roscher, this view of history as an art form gradually lost ground to an understanding of history as a science; that is, as the critical, rational and empirically grounded study of the past (Muhlack 2011Muhlack, Ulrich 2011 “Herodotus and Thucydides in the View of Nineteenth-Century German Historians.” Translated by Neville Morley. In The Western Time of Ancient History: Historiographical Encounters with the Greek and Roman Pasts, edited by Alexandra Lianeri, 179–209. Cambridge: Cambridge University Press. , 181; Pires 2006Pires, Francisco Murari 2006 “Thucydidean Modernities: History between Science and Art.” In Brill’s Companion to Thucydides, edited by Antonis Tsakmakis and Antonios Rengakos, 811–837. Leiden: Brill. , 811; White 1973White, Hayden 1973 “Interpretation in History.” New Literary History 4 (2): 281–314. , 283). No longer was historiography principally concerned with the construction of a clear and instructive narrative, but instead with the gathering of accurate information and the strict presentation of facts in order to ascertain “wie es eigentlich gewesen” (‘how it really was’) (Ranke 1824Ranke, Leopold 1824 Geschichten der romanischen und germanischen Völker von 1494 bis 1535. Leipzig: G. Reimer., vi). As Collingwood (1946Collingwood, Robin 1946 The Idea of History. Oxford: Oxford University Press., 126) has discussed, this shift had partly to do with a desire to distinguish history from philology, theology and moral philosophy, and to establish it as an institutionalised scholarly discipline in its own right; but it also came from the growing influence of scientific positivism across all intellectual fields of study (Harloe and Morley 2012Harloe, Katharine, and Neville Morley 2012 “Introduction: The Modern Reception of Thucydides.” In Thucydides and the Modern World: Reception, Re-interpretation and Influence from the Renaissance to the Present, edited by Katharine Harloe and Neville Morley, 1–24. Cambridge: Cambridge University Press. , 9; Lianeri 2015Lianeri, Alexandra 2015 “On Historical Time and Method: Thucydides’ Contemporary History in Nineteenth-Century Britain.” In A Handbook to the Reception of Thucydides, edited by Christine Lee and Neville Morley, 176–196. Chichester: Wiley., 177; Morley 2014Morley, Neville 2014 Thucydides and the Idea of History. London: I. B. Tauris. , 43). Roscher ([1854] 1868Roscher, Wilhelm (1854) 1868 System der Volkswirthschaft, ein Hand- und Lesebuch für Geschäftsmänner und Studierende, Band 1. 5th ed. Stuttgart: Verlag der I. G. Cottaschen Buchhandlung., 45) in particular was keen to draw explicit parallels between the work of the historian and the work of the scientist, writing, for example, that the historical method adopted in his study of The Principles of Political Economy proceeds “wie die Naturforscher” (‘in the manner of the naturalist’). He suggested that historians too should engage in the “mikroskopischen Untersuchungen” and “Sectionen” (‘microscopic examinations and dissections’) of the past, and that his task in his book was no more than detailing changes in “die Anatomie und Physiologie” (‘the anatomy and physiology’) of societies through time (45).

The present study aims to explore how this shift in attitudes towards the proper aims and methods of history writing might have shaped the interpretation, translation and reception of Thucydides’ History of the Peloponnesian War in Britain. As the next section seeks to explain, Thucydides’ History has been one of the most frequently retranslated texts in the classical Greek canon and, along with the writings of Herodotus, it is widely regarded as one of the founding works of the Western discipline of historiography. Consequently, this paper attempts to examine whether and how this ancient text was translated differently by English-language translators working in the nineteenth century (Bloomfield 1829Bloomfield, Samuel 1829 The History of Thucydides, Newly Translated into English. London: Longman, Rees, Orme, Brown and Green.; Dale 1848Dale, Henry 1848 The History of the Peloponnesian War by Thucydides: A New and Literal Version. London: George Bell and Sons.; Crawley 1874Crawley, Richard 1874 History of the Peloponnesian War by Thucydides. London: Longmans, Green and Co.; Jowett 1881Jowett, Benjamin 1881 Thucydides, Translated into English. 2 vols. Oxford: Clarendon Press.) in comparison with those renderings produced of this work in the seventeenth and eighteenth centuries (Hobbes [1629] 1843Hobbes, Thomas (1629) 1843 “The Peloponnesian War.” In The English Works of Thomas Hobbes of Malmesbury, Vols. VIII–XI, edited by William Molesworth. London: John Bohn.; Smith 1753Smith, William 1753 The History of the Peloponnesian War. 2 vols. London: John Watts.). This analysis is carried out by means of a corpus-based methodology which, I argue, can better enable researchers to engage with each translator’s overall presentation of the source text through the production and interrogation of concordances listing every instance of a given search item as it occurs within digitised versions of the translated works. This is demonstrated through an investigation into the translators’ uses of the term ‘fact(s)’ which reveals a striking divergence in interpretation between the six translations: Crawley’s (1874)Crawley, Richard 1874 History of the Peloponnesian War by Thucydides. London: Longmans, Green and Co. History in particular appears to lend a significantly more objective and empirical tone to Thucydides in English.

2.Thucydides’ History

Thucydides’ History of the Peloponnesian War is a key text in the classical Greek canon. Its author was an Athenian general, stationed on the island of Thasos when conflict first broke out against the Spartans in 431 BCE (Lattimore 1998Lattimore, Stephen 1998 Thucydides: The Peloponnesian War. Indianapolis, IN: Hackett., xvi). Despite his military rank, however, Thucydides’ active participation in the fighting appears to have been only fleeting: in 424 BCE he was sent into exile by the Athenian polis in punishment for losing a decisive battle. Although clearly hurt by such a harsh sanction, this did mean that Thucydides was able to move with considerable freedom around the war zone and to associate with soldiers fighting both for and against Athens. As he wrote in Book 5 of his History, “being present at the activities of both sides, especially the Peloponnesians, unoccupied because of my exile, I understood these all the more” (5.26.5 – trans. Lattimore 1998Lattimore, Stephen 1998 Thucydides: The Peloponnesian War. Indianapolis, IN: Hackett.). This apparent desire to report multiple points of view and to provide a more balanced account of the events is what has since earned him fame as the ‘founder’ of critical scientific historiography (Morley 2016Morley, Neville 2016 “The Anti-Thucydides: Herodotus and the Development of Modern Historiography.” In Brill’s Companion to the Reception of Herodotus in Antiquity and Beyond, edited by Jessica Priestley and Vasiliki Zali, 143–166. Leiden: Brill. , 208). Indeed, Morley (2014)Morley, Neville 2014 Thucydides and the Idea of History. London: I. B. Tauris. , Muhlack (2011)Muhlack, Ulrich 2011 “Herodotus and Thucydides in the View of Nineteenth-Century German Historians.” Translated by Neville Morley. In The Western Time of Ancient History: Historiographical Encounters with the Greek and Roman Pasts, edited by Alexandra Lianeri, 179–209. Cambridge: Cambridge University Press. , Pires (2006)Pires, Francisco Murari 2006 “Thucydidean Modernities: History between Science and Art.” In Brill’s Companion to Thucydides, edited by Antonis Tsakmakis and Antonios Rengakos, 811–837. Leiden: Brill. and Süßmann (2012)Süßmann, Johannes 2012 “Historicising the Classics: How Nineteenth-Century German Historiography Changed the Perspective on Historical Tradition.” In Thucydides and the Modern World: Reception, Re-interpretation and Influence from the Renaissance to the Present, edited by Katharine Harloe and Neville Morley, 77–92. Cambridge: Cambridge University Press. have all demonstrated the extent to which Thucydides’ History has frequently been held up as the supreme methodological model for historians throughout the modern period, as someone who could serve as a prime example for the “study and writing of history itself” (Harloe and Morley 2012Harloe, Katharine, and Neville Morley 2012 “Introduction: The Modern Reception of Thucydides.” In Thucydides and the Modern World: Reception, Re-interpretation and Influence from the Renaissance to the Present, edited by Katharine Harloe and Neville Morley, 1–24. Cambridge: Cambridge University Press. , 9).

At least partly for this reason, Thucydides has been repeatedly translated by successive generations of translators, each working in quite different historical, cultural and political contexts across Europe from the Renaissance onwards. The History was first made available to English-speaking readers via Thomas Nicholls’ (1550)Nicholls, Thomas 1550 The Hystory Writtone by Thucidides the Athenyan of the Warre, whiche was betwene the Peloponesians and the Athenyans, Translated oute of Frenche into the Englysh Language by Thomas Nicolls Citezeine and Goldesmyth of London. London: William Tylle. translation; this, however, relied exclusively on Claude de Seyssel’s French-language rendering, rather than the original Greek, and so it is Thomas Hobbes’ ([1629] 1843)Hobbes, Thomas (1629) 1843 “The Peloponnesian War.” In The English Works of Thomas Hobbes of Malmesbury, Vols. VIII–XI, edited by William Molesworth. London: John Bohn. text which is most widely regarded as the first true translation of Thucydides’ masterpiece into English. Hobbes’ version has since been followed by at least ten full retranslations over the past four hundred years (Smith 1753Smith, William 1753 The History of the Peloponnesian War. 2 vols. London: John Watts.; Bloomfield 1829Bloomfield, Samuel 1829 The History of Thucydides, Newly Translated into English. London: Longman, Rees, Orme, Brown and Green.; Dale 1848Dale, Henry 1848 The History of the Peloponnesian War by Thucydides: A New and Literal Version. London: George Bell and Sons.; Crawley 1874Crawley, Richard 1874 History of the Peloponnesian War by Thucydides. London: Longmans, Green and Co.; Jowett 1881Jowett, Benjamin 1881 Thucydides, Translated into English. 2 vols. Oxford: Clarendon Press.; Smith 1919–1923Smith, Charles Forster 1919–1923 Thucydides. Cambridge, MA: Harvard University Press.; Warner 1954Warner, Rex 1954 Thucydides: The Peloponnesian War. Harmondsworth: Penguin.; Blanco 1998Blanco, Walter 1998 The Peloponnesian War. London: Norton.; Lattimore 1998Lattimore, Stephen 1998 Thucydides: The Peloponnesian War. Indianapolis, IN: Hackett.; Hammond 2009Hammond, Martin 2009 Thucydides: The Peloponnesian War. Oxford: Oxford University Press.), as well as a number of partial retranslations that have focused on specific sections of the source (e.g., Henry Musgrave Wilkins’ Speeches from Thucydides, published in 1873Wilkins, Henry Musgrave 1873 Speeches from Thucydides. Translated into English for the Use of Students, with an Introduction and Notes. 2nd ed. London: Longmans, Green and Co.). Taken together, it is argued that this collection of successive translations constitutes an invaluable resource with which to observe and investigate the ways in which modern interpretations of ancient historical texts have been shaped by ongoing debates surrounding the proper aims and methods of good history writing.

3.Translating Thucydides

While Thucydides has for many years now been the focus of a sustained research effort by scholars working in the fields of classics and classical reception studies, issues pertaining to the translation of this ancient author remain substantially under-explored (Greenwood 2015Greenwood, Emily 2015 “On Translating Thucydides.” In A Handbook to the Reception of Thucydides, edited by Christine Lee and Neville Morley, 91–122. Chichester: Wiley., 91). Greenwood (2015Greenwood, Emily 2015 “On Translating Thucydides.” In A Handbook to the Reception of Thucydides, edited by Christine Lee and Neville Morley, 91–122. Chichester: Wiley., 102–107) has provided perhaps the most detailed comparison of the different English translations of Thucydides through her discussion of the multiple renderings produced of one of the best known passages in the History (1.20.1–1.22.4), and especially how (un)successful they have each been in reproducing the central metaphor it contains of historiography as physical exertion. Her analysis is made, Greenwood (2015Greenwood, Emily 2015 “On Translating Thucydides.” In A Handbook to the Reception of Thucydides, edited by Christine Lee and Neville Morley, 91–122. Chichester: Wiley., 101–104) asserts, to highlight the influential role of the translator as one who “effectively controls the parameters of what Thucydides can mean” for readers without access to the Greek, and thus the “difference that translation makes” in mediating Thucydides’ own presentation of his credentials as a historian for most members of the receiving culture.

Harloe and Morley (2012Harloe, Katharine, and Neville Morley 2012 “Introduction: The Modern Reception of Thucydides.” In Thucydides and the Modern World: Reception, Re-interpretation and Influence from the Renaissance to the Present, edited by Katharine Harloe and Neville Morley, 1–24. Cambridge: Cambridge University Press. , 14–15) have similarly discussed a variety of English translations of an important phrase in this same section of the work, “κατὰ τὸ ἀνθρώπινον” (1.22.4), again arguing that the way in which this is interpreted is “essential for understanding Thucydides’ overall views of both history and of historiography.” As they explain, these words are used by Thucydides as part of his claim that his text will be valuable to future readers because, “according to the human thing” (to translate the Greek as literally as possible), history tends to repeat itself. Different English translators have rendered this as “according to human nature,” “the human condition,” “the human situation,” “humans being what they are” or even “the world being the way it is”: as Harloe and Morley (2012Harloe, Katharine, and Neville Morley 2012 “Introduction: The Modern Reception of Thucydides.” In Thucydides and the Modern World: Reception, Re-interpretation and Influence from the Renaissance to the Present, edited by Katharine Harloe and Neville Morley, 1–24. Cambridge: Cambridge University Press. , 14) note, the subtle differences between each of these solutions hold major implications for the extent to which we can frame Thucydides as an ancient historian whose writings deliberately sought to reveal universal, transhistorical principles of social behaviour.11. Alexandra Lianeri (2002)Lianeri, Alexandra 2002 “Translation and the Establishment of Liberal Democracy in Nineteenth-Century England: Constructing the Political as an Interpretative Act.” In Translation and Power, edited by Maria Tymoczko and Edwin Gentzler, 1–24. Amherst, MA: University of Massachusetts Press. too has explored the translation of Thucydides. However, her analysis focuses mainly on how changes in the political climate of the receiving context have affected the interpretation of the History, rather than investigating the impact of developments in historiography.

Both of these analyses thus highlight the importance of translation in shaping modern understandings of and interest in classical history and historiography (Greenwood 2015Greenwood, Emily 2015 “On Translating Thucydides.” In A Handbook to the Reception of Thucydides, edited by Christine Lee and Neville Morley, 91–122. Chichester: Wiley., 91; Harloe and Morley 2012Harloe, Katharine, and Neville Morley 2012 “Introduction: The Modern Reception of Thucydides.” In Thucydides and the Modern World: Reception, Re-interpretation and Influence from the Renaissance to the Present, edited by Katharine Harloe and Neville Morley, 1–24. Cambridge: Cambridge University Press. , 14). Yet, so far missing from such discussion of the Thucydidean translations is investigation into the ways in which the Greek author’s English translators might have responded to the ‘scientisation’ of history as a field of intellectual activity and any consequent shifts in historiographical discourse. It is to this analysis that we shall now turn.

4.“Looked at by the light of facts”

A useful starting point for our investigation can be identified by comparing the variety of English renderings produced in the seventeenth, eighteenth and nineteenth centuries of the section of Thucydides’ text found at 5.26.2. This passage is often known as the ‘Second Preface’ of the History, given that it is in this paragraph that the author steps back from his description of the events of the war to argue that the so-called ‘Peace of Nicias’ was not in truth a break in the hostilities between Athens and Sparta, and consequently that the conflict described in the second half of the History should be considered a continuation of that already presented. The Greek text is copied in Example (1a),22.As is made clear in their prefaces, each of the six translators has relied on a different edition of Thucydides’ Greek: for example, Hobbes used Aemilius Portus’ (1594)Portus, Aemilius ed. 1594 Thucydidis Olori Filii, De Bello Peloponnesiaco Libri Octo. Frankfurt: Andrea Wecheli. version of the source, while Dale depended on Arnold’s (1828–1835)Arnold, Thomas ed. 1828–1835 The History of the Peloponnesian War, by Thucydides. Oxford: John Henry Parker. and Jowett on Poppo’s (1843–1851)Poppo, Ernst Friedrich ed. 1843–1851 Thucydidis De bello Peloponnesiaco libri octo. Gothae: Sumtibus Fridericae Hennings.. That said, I have checked the wording of this passage in each of the relevant editions and there appears to be no variation at this point in the text. followed by the translations by Hobbes ([1629] 1843)Hobbes, Thomas (1629) 1843 “The Peloponnesian War.” In The English Works of Thomas Hobbes of Malmesbury, Vols. VIII–XI, edited by William Molesworth. London: John Bohn., Smith (1753)Smith, William 1753 The History of the Peloponnesian War. 2 vols. London: John Watts., Bloomfield (1829)Bloomfield, Samuel 1829 The History of Thucydides, Newly Translated into English. London: Longman, Rees, Orme, Brown and Green., Dale (1848)Dale, Henry 1848 The History of the Peloponnesian War by Thucydides: A New and Literal Version. London: George Bell and Sons., Crawley (1874)Crawley, Richard 1874 History of the Peloponnesian War by Thucydides. London: Longmans, Green and Co. and Jowett (1881)Jowett, Benjamin 1881 Thucydides, Translated into English. 2 vols. Oxford: Clarendon Press..

Based on the evidence of this one example alone, it would appear that an important shift did occur at some point during the mid-nineteenth century in the interpretation of Thucydides and in the target-culture presentation of his conception of historical truth. Hobbes, Smith and Bloomfield all seem to have interpreted this passage in a way that asks the reader only to look into the “actions” or “transactions” (τοῖς ἔργοις) of the past as they are “narrated,” “related” or “set down” (διῄρηται) in the History itself. In other words, these earlier renderings suggest that the truth of the matter can be understood on the basis of evidence found within Thucydides’ account of the war, and no reference is made to sources of truth existing externally from this narrative. We are asked simply to assess the internal coherence of the story told, and not to consider its relationship to any objective reality. Dale, Crawley and Jowett, on the other hand, all reject this previous interpretation and imply that ‘the facts of the case’ exist independently of the Greek historian’s text; they indicate that these facts have a reality and authority of their own, and that it is these objective sources of truth that should be examined when discussing the Peace of Nicias. Indeed, in Crawley’s version especially, the clear implication is that Thucydides himself has “rationally considered” the facts in order to arrive at his conclusion that the entire period of 431–404 BCE should be dealt with as one continuous war. In this way, the three later translators’ word choices – and Crawley’s in particular – appear to lend a much greater degree of empirical objectivity to the tone of Thucydides in English at this point in the text. This tone could be seen as more appropriate to the new historiographical ideals circulating in their nineteenth-century Britain, as well as to Thucydides’ growing status in this receiving context as a model scientific historian.

That said, and despite the apparent strength of this difference between the six interpretations, it would be difficult to support any attempt to make more general claims about each translator’s overall presentation of Thucydides on the basis of their renderings of this one passage. After all, 5.26.2 is likely only one of several hundreds of extracts of potential relevance to our interests here. For this reason, the following sections attempt to show the potential of a corpus-based approach as a means of exploring the extent to which this observed difference between the translations can be said to be valid across each translator’s rendering of Thucydides’ text as a whole. This paper thus additionally aims to contribute methodologically to the study of the modern reception and translation of Thucydides by demonstrating the potential of digital corpus analysis tools for this area of investigation.33.I use ‘corpus-based methodology’ here somewhat differently to the way in which this approach is typically discussed, for example, in the field of linguistics. In this study, my corpus does not consist of a “large, representative sample of a particular type of naturally occurring language” which we might use to make claims about a range of linguistic phenomena (Baker 2006Baker, Paul 2006 Using Corpora in Discourse Studies. London: Continuum., 2). Rather, as in literary studies such as Balossi (2014)Balossi, Giuseppina 2014  A Corpus Linguistic Approach to Literary Language and Characterisation: Virginia Woolf’s The Waves. Amsterdam: John Benjamins. and Hubbard (2002)Hubbard, E. Hilton 2002 “Conversation, Characterisation and Corpus Linguistics: Dialogue in Jane Austen’s Sense and Sensibility .” Literator 23 (2): 67–85. , ‘corpus’ is used here to refer to electronically encoded versions of the six texts in my dataset (the English-language translations of Thucydides’ History); my ‘corpus-based methodology’ consists therefore in the application of a suite of specially developed computer programmes to produce both quantitative and qualitative insight into the differences between these renderings of the source.

5.A corpus-based methodology

While to my knowledge corpus-based approaches have not previously been applied to the study of the modern reception of Thucydides, the use of corpus concordancing tools does now constitute a well-established methodology within translation studies more generally (Bernardini and Kenny 2020Bernardini, Silvia, and Dorothy Kenny 2020 “Corpora.” In Routledge Encyclopedia of Translation Studies. 3rd ed., edited by Mona Baker and Gabriela Saldanha, 110–115. London: Routledge.). Here, corpora have proved invaluable in allowing translation scholars to interrogate large collections of translated text and answer a variety of research questions; for example, they have been used to identify otherwise invisible patterns specific to translated English as opposed to monolingual writing (Baker 1993Baker, Mona 1993 “Corpus Linguistics and Translation Studies: Implications and Applications.” In Text and Technology: In Honour of John Sinclair, edited by Mona Baker, Gill Francis, and Elena Tognini-Bonelli, 233–252. Amsterdam: John Benjamins. ; Olohan 2003Olohan, Maeve 2003 “How Frequent Are the Contractions? A Study of Contracted Forms in the Translational English Corpus.” Target 15 (1): 59–89. ; Olohan and Baker 2000Olohan, Maeve, and Mona Baker 2000 “Reporting That in Translated English: Evidence for Subconscious Processes of Explicitation?” Across Languages and Cultures 1 (2): 141–158. ; Tirkkonen-Condit 2004Tirkkonen-Condit, Sonja 2004 “Unique Items – Over- or Under-Represented in Translated Language?” In Translation Universals: Do They Exist?, edited by Anna Mauranen and Pekka Kujamäki, 177–186. Amsterdam: John Benjamins. ), to explore the nature of a particular translator’s style (Baker 2000Baker, Mona 2000 “Towards a Methodology for Investigating the Style of a Literary Translator.” Target 12 (2): 241–266. ), and to investigate the impact of translators’ choices on narratological structures of foreign works of literature (Bosseaux 2007Bosseaux, Charlotte 2007 How Does It Feel? Point of View in Translation: The Case of Virginia Woolf into French. Amsterdam: Rodopi. ).

The Genealogies of Knowledge project,44.Project website: www​.genealogiesofknowledge​.net/ of which the present study forms a part, represents one of the most ambitious ventures in corpus-based translation studies to date. Having started in 2016 at the University of Manchester (UK), it is building a multilingual set of corpora to investigate the role of translation and other sites of remediation in the evolution and contestation of political and scientific discourse from antiquity through to the present day. This set currently includes collections of original and translated texts written in Classical Greek, Latin and Medieval Arabic – chosen as key historical languages for the production and transmission of knowledge – and a ‘Modern’ corpus of English-language translations and commentaries, focused for the most part on the (re)interpretation of Classical Greek, Latin and Arabic works produced in the nineteenth and twentieth centuries.55.The project is also developing a corpus of Internet discourse which aims to help researchers explore how concepts such as ‘democracy’, ‘truth’ and ‘expertise’ are now being contested and renegotiated online by radical activist groups and alternative media outlets at both ends of the political spectrum. Indeed, alongside English versions of key works by Plato, Aristotle, Herodotus, Hippocrates, Galen, Cicero, Tacitus, Livy and Averroes, twelve different translations of (sections of) Thucydides’ History have so far been added to the electronic corpus. These include the six target texts on which the present study focuses, as well as a further two full translations produced later in the twentieth century (Smith 1919–1923Smith, Charles Forster 1919–1923 Thucydides. Cambridge, MA: Harvard University Press.; Lattimore 1998Lattimore, Stephen 1998 Thucydides: The Peloponnesian War. Indianapolis, IN: Hackett.) and a number of partial translations (e.g., Wilkins 1873Wilkins, Henry Musgrave 1873 Speeches from Thucydides. Translated into English for the Use of Students, with an Introduction and Notes. 2nd ed. London: Longmans, Green and Co.).66.We have not unfortunately been able to secure copyright permission to include Warner’s 1954Warner, Rex 1954 Thucydides: The Peloponnesian War. Harmondsworth: Penguin. translation or Hammond’s 2009Hammond, Martin 2009 Thucydides: The Peloponnesian War. Oxford: Oxford University Press. version in our project corpus.

Within corpus-based translation studies, analysis may begin simply by generating a concordance displaying all instances of a set of search items as they appear within every text under study, and noting their frequency in each. As demonstrated below, this purely quantitative mode of investigation can expose intriguing differences between translations in terms of their translators’ relative preference for certain kinds of terminology, as well as suggesting important “focal words” to be investigated in more detail through qualitative analysis (Kempannen 2004Kempannen, Hannu 2004 “Keywords and Ideology in Translated History Texts: A Corpus-Based Analysis.” Across Languages and Cultures 5 (1): 89–106. , 91). In line with Paul Baker’s (2006Baker, Paul 2006 Using Corpora in Discourse Studies. London: Continuum., 49) suggestions for using corpora in discourse studies, we might also usefully take into account at this point the “dispersion” of particular tokens within each text; in other words, whether occurrences of a term are all clumped together in one section of the translation being examined, or whether “the word is a constant feature, cropping up every now and again with regularity.” Such analyses can thus help to determine not only how frequently a given term is deployed, but where in each text it appears most commonly, for example, in the translator’s introduction, in endnotes, or in the main body of the translation.77.The suite of corpus analysis tools being developed for the Genealogies of Knowledge project does not currently include visualisation software capable of indicating the dispersion of a concept within any given translation. Nevertheless, research using the Genealogies of Knowledge corpus browser can be supplemented with freeware such as AntConc’s ‘Concordance Plot’, available for download at http://​www​.laurenceanthony​.net​/software​/antconc/. This displays a linear graph showing the location of every instance of the search term from the beginning to the end of the text and thus provides an invaluable starting point from which to understand the contexts in which this word is used by each translator.

Revealing insights are often produced by examining patterns of collocation, defined by Firth ([1957] 1968Firth, John (1957) 1968 “A Synopsis of Linguistic Theory, 1930–55.” In Selected Papers of J. R. Firth 1952–1959, edited by Frank Palmer, 168–205. London: Longman., 182) and Sinclair (1991Sinclair, John 1991 Corpus, Concordance, Collocation. Oxford: Oxford University Press., 170) as the recurrent tendency of two or more words to appear in close proximity to each other. Such patterns can easily be identified by using the corpus browser to generate a concordance of all instances of the search term (e.g., ‘facts’ in Figure 1) as they occur in each translation and by sorting the co-text displayed in a concordance alphabetically either to the left or the right of this node word. This process can frequently uncover divergent uses of a particular lexical item (or cluster of items) by different translators, for example, the propensity to modify a noun with a specific kind of adjective. These patterns are especially interesting and significant if we follow the Firthian argument that “words (or signifiers to use Saussure’s term [1974Saussure, Ferdinand de 1974 Course in General Linguistics. Translated by Wade Baskin. Revised ed. London: Fontana.]) can only take on meaning (that which is signified) by the context that they occur in” (Baker 2006Baker, Paul 2006 Using Corpora in Discourse Studies. London: Continuum., 96): through the analysis of these collocates, it is suggested, we can fruitfully identify many of the associations, connotations and assumptions embodied in a translator’s use of a lexical item. Or, to put it another way, by investigating, for example, the adjectives, nouns, verbs, conjunctions, and prepositions that frequently co-occur next to the chosen node word we can begin to uncover what each translator understands this term to mean and to study the discourses that underpin its deployment in each target text.

6.Frequency analysis

The key difference between the translations highlighted in my analysis of 5.26.2 above had to do with the three later translators’ presentation of ‘facts’ as external and objective sources of truth. Their use of this term is especially interesting because it is derived from no direct equivalent in the (classical) Greek language (Wootton 2015Wootton, David 2015 The Invention of Science: A New History of the Scientific Revolution. London: Allen Lane., 254), and because the development of the modern ‘fact’ has widely been discussed as a key invention in the history of scientific discourse, shaping debates over the aims and methods of the natural sciences from the seventeenth century onwards (Poovey 1998Poovey, Mary 1998 A History of the Modern Fact: Problems of Knowledge in the Sciences of Wealth and Society. Chicago, IL: University of Chicago Press. ; Shapiro 2000Shapiro, Barbara 2000 A Culture of Fact: England, 1550–1720. Ithaca, NY: Cornell University Press.). As a consequence, it makes sense to focus our corpus-based analysis on the use of this term – in both its singular and plural forms – in each of the six translations. Indeed, by using the software to identify all instances of ‘fact’ and ‘facts’ as they appear throughout these English renderings of Thucydides’ History, we find an intriguing disparity in the relative use of these lexical items. Table 1 shows the results of this analysis: in order to highlight patterns in dispersion, those instances of the search terms that are found in the various paratextual elements of each text are listed separately (under the columns labelled ‘PT’ for ‘Paratext’) from those extracted from the main body of the translations.

Table 1.Frequency of ‘fact’ and ‘facts’ in each of the six English translations of Thucydides’ History

Translator	Date	‘fact’			‘facts’			Total
		PT	TT		PT	TT		PT	TT
Hobbes	(1629) 1843Hobbes, Thomas (1629) 1843 “The Peloponnesian War.” In The English Works of Thomas Hobbes of Malmesbury, Vols. VIII–XI, edited by William Molesworth. London: John Bohn.	1	18		0	3		1	21
Smith	1753Smith, William 1753 The History of the Peloponnesian War. 2 vols. London: John Watts.	11	45		10	7		21	52
Bloomfield	1829Bloomfield, Samuel 1829 The History of Thucydides, Newly Translated into English. London: Longman, Rees, Orme, Brown and Green.	1	23		2	3		3	26
Dale	1848	10	43		1	11		11	54
Crawley	1874Crawley, Richard 1874 History of the Peloponnesian War by Thucydides. London: Longmans, Green and Co.	0	78		0	20		0	98
Jowett	1881Jowett, Benjamin 1881 Thucydides, Translated into English. 2 vols. Oxford: Clarendon Press.	0	24		5	10		5	34

Note. PT = Paratext (introductions, prefaces, footnotes); TT = Target text.

As Table 1 shows, Crawley’s (1874)Crawley, Richard 1874 History of the Peloponnesian War by Thucydides. London: Longmans, Green and Co. translation employs the term ‘fact(s)’ much more frequently than those of his predecessors and even more than those of his more contemporary rivals (i.e., Jowett). While Hobbes ([1629] 1843)Hobbes, Thomas (1629) 1843 “The Peloponnesian War.” In The English Works of Thomas Hobbes of Malmesbury, Vols. VIII–XI, edited by William Molesworth. London: John Bohn., for example, uses these items just 21 times in the main body of his translation (as opposed to in his introduction and/or footnotes), ‘fact’ and ‘facts’ can be found 98 times throughout Crawley’s text. The differences in these figures are significant according to a statistical measure of significance commonly used in corpus linguistics: performing a log likelihood test comparing the number of occurrences of ‘fact’ and ‘facts’ in Crawley’s target text with the combined number of occurrences in Hobbes’, Smith’s, Bloomfield’s, Dale’s and Jowett’s translations gives us a score of 71.91.88.This calculation was made using the online calculator freely available on the University of Lancaster’s UCREL website: http://​ucrel​.lancs​.ac​.uk​/llwizard​.html. The combined wordcount for Hobbes’, Smith’s, Bloomfield’s, Dale’s and Jowett’s translations is 1,241,325 tokens, while for Crawley’s the total number of tokens is 205,064. For more information on the use of log-likelihood scores as a measure of statistical significance, see Gabrielatos (2018)Gabrielatos, Costas 2018 “Keyness Analysis: Nature, Metrics and Techniques.” In Corpus Approaches to Discourse: A Critical Review, edited by Charlotte Taylor and Anna Marchi, 225–258. Oxford: Routledge. . This corresponds to a Bayes factor of 57.72, which is dramatically higher than the conventional cut-off of 10, meaning we can be highly confident that this variation in the use of ‘fact’ and ‘facts’ is not due simply to chance (Gabrielatos 2018Gabrielatos, Costas 2018 “Keyness Analysis: Nature, Metrics and Techniques.” In Corpus Approaches to Discourse: A Critical Review, edited by Charlotte Taylor and Anna Marchi, 225–258. Oxford: Routledge. , 240). This initial analysis would therefore suggest that, in its emphasis on ‘facts’ as the basis of Thucydides’ History, Crawley’s translation may stand out from all previous versions as well as from its more or less contemporary rivals.

7.Concordance analysis

7.1 Hobbes ([1629] 1843)Hobbes, Thomas (1629) 1843 “The Peloponnesian War.” In The English Works of Thomas Hobbes of Malmesbury, Vols. VIII–XI, edited by William Molesworth. London: John Bohn. 99.The edition of Hobbes’ translation included in the Genealogies of Knowledge corpus is that produced by William Molesworth in 1843. This edition modernised the spelling of Hobbes’ text but otherwise appears not to differ in any significant respect from the original publication. The lines cited above have nevertheless been double-checked against the 1629 version to ensure no important changes were introduced by the nineteenth-century editor.

Quantitative data of the kind presented in Section 6 can only ever produce relatively crude insights into the differences of interpretation expressed by each translator. Qualitative inspection of the individual concordance lines, on the other hand, allows us to explore these texts in much more detail. As noted above, this process is greatly facilitated by the concordance browser software which enables the user to sort the lines to the right and left of the search term. Indeed, starting with the use of ‘fact’ and ‘facts’ in Thomas Hobbes’ version, we find that in every case these items refer to actions performed by the protagonists of Thucydides’ narrative. To use Wootton’s (2015Wootton, David 2015 The Invention of Science: A New History of the Scientific Revolution. London: Allen Lane., 252) terminology, they reflect an “agency idea of facts,” indicating “something that has occurred because someone has done it.” As can be seen in the four concordance lines in Example (2), this meaning is suggested most obviously through the repeated presence of dynamic verbs (“done,” “begin,” “committed” and “innovate”) and of a named agent (“the women,” “they,” “they” and “you”) within the immediate co-text to either side of the search words.

(2)

1. fact of the women concerning Itys was done there

2. And they were to begin the fact themselves

3. And together with these, they committed other facts of the same kind

4. you did not by us as we did by you, first innovate nothing in fact and then with words persuade us to go forth again

Even in cases where no agent is specified explicitly, the implication remains that what is being referred to are actions committed by someone. In the concordance lines in Example (3), this is most often evident through the opposition in Hobbes’ text of ‘facts’ to rhetoric; a ‘fact’ for this translator is what is done as opposed to what is said.

(3)

1. by presenting (unto these men) a trial not of words but of facts which, if they be good, a short narration of them will serve the turn

2. I, according to the prescript of the law, delivered in word what was expedient; and those that are here interred have in fact been already been honoured

3. It was in name a state democratical, but in fact a government of the principal man

In each of the examples cited so far, ‘facts’ are presented as key components in the structure of Thucydides’ narrative: they are essentially understood as the events and deeds which the History seeks to relate. These ‘facts’ are presented by Hobbes’ Thucydides as having actually occurred and their truth is never held in doubt. Yet, their validity as descriptions of reality is firmly rooted in and dependent on the subjective experience and representation of the text’s narrator (or one of the other speakers to whom Thucydides’ temporarily gives the role of narrator). They are not ‘brute’ facts, existing independently from the observer and his narrative (see Wootton 2015Wootton, David 2015 The Invention of Science: A New History of the Scientific Revolution. London: Allen Lane., 256); they are not sources of truth positioned externally from the text, nor are they assumed to hold unparalleled authority of their own in settling debates around what actually happened.

This finding is not surprising when we take into consideration the early modern context in which Hobbes was writing. As Wootton (2015Wootton, David 2015 The Invention of Science: A New History of the Scientific Revolution. London: Allen Lane., 253) has suggested, the modern usage of ‘fact’ to mean a type of knowledge grounded in experience did not appear in England before the late 1650s. Derived from the Latin verb facere (‘to do’) – or to be precise its neuter past participle factum – a ‘fact’ in Hobbes’ England referred simply to “that which has been done” and was used primarily in legal settings to refer to deeds and crimes committed (Shapiro 2000Shapiro, Barbara 2000 A Culture of Fact: England, 1550–1720. Ithaca, NY: Cornell University Press., 228; Wootton 2015Wootton, David 2015 The Invention of Science: A New History of the Scientific Revolution. London: Allen Lane., 283–284). This former usage can clearly be seen in other corpora of other texts produced in the 1620s and 1630s. Example (4) shows lines retrieved from the Early English Books Online corpus developed by the SAMUELS project,1010.Available at: https://​www​.english​-corpora​.org​/eebo/ (last accessed 05/04/2019). in which ‘fact’ is clearly being used to refer to specific actions with specific agents, in much the same way as we find in Hobbes’ translation of Thucydides.

(4)

1. being examined concerning this bloudy fact, he plainly confessed (Boccaccio, The Decameron, trans. Florio, 1620)

2. being assured that they the fathers wil not condemne any mans fact before they know it (Sarpi, The Historie of the Councel of Trent, trans. Brent, 1629)

3. the pope had induced Peter de Vineis to undertake this foule fact (Sutcliffe, A True Relation of Englands Happinesse, Under the Raigne of Queene Elizabeth and the Miserable Estate of Papists, Under the Popes Tyranny, 1629)

Nevertheless, this patterning in Hobbes’ text is interesting when placed in contradistinction with the more scientific usage of ‘fact’ and ‘facts’ that can be found deployed in later translations of Thucydides. As we will see, Dale, Crawley and Jowett all repeatedly make use of ‘facts’ in the sense of a form of knowledge universally held to be true, and having a natural authority and agency of its own; these later translators all take care to emphasise these ‘facts’ as the crucial pieces of independent, objective evidence out of which Thucydides’ History has been constructed.

7.2 Smith (1753)Smith, William 1753 The History of the Peloponnesian War. 2 vols. London: John Watts.

Before we examine these three later translations, it is interesting to observe the patternings present in William Smith’s (1753)Smith, William 1753 The History of the Peloponnesian War. 2 vols. London: John Watts. and Samuel Bloomfield’s (1829)Bloomfield, Samuel 1829 The History of Thucydides, Newly Translated into English. London: Longman, Rees, Orme, Brown and Green. versions. As mentioned above, the scholarly literature on the history of the scientific fact in England generally locates the emergence of this discourse in the late 1650s and early 1660s. According to Wootton (2015Wootton, David 2015 The Invention of Science: A New History of the Scientific Revolution. London: Allen Lane., 293), it was at this time that the modern sense of ‘fact’ not only suddenly became “linguistically commonplace” in scientific reports, but also “institutionally entrenched” in the statutes of the newly established Royal Society. It is therefore notable that in Smith’s translation, produced nearly 100 years after this revolution in the natural sciences, we find that ‘fact’ and ‘facts’ continue in the overwhelming majority of cases to be associated with specific actions and agents, as was the case in Hobbes’ rendering. Here too, as shown in Example (5), ‘facts’ are primarily found associated with dynamic verbs (“assisted at”), equated with crimes such as “trespass,” and/or directly opposed to rhetoric (“words”).

(5)

1. and a stranger from Argos, who assisted at the fact, being apprehended and tortured

2. But the fact was different in the light they saw it, and trespass had not been committed

3. They are only of weight, when facts and words are equitably to be judged

In contrast with Hobbes’ translation, however, there is some evidence that, already in Smith’s 1753Smith, William 1753 The History of the Peloponnesian War. 2 vols. London: John Watts. version, facts were beginning to lose their exclusive association with the particular deeds of named individuals, a finding which would concur with Shapiro’s (2000Shapiro, Barbara 2000 A Culture of Fact: England, 1550–1720. Ithaca, NY: Cornell University Press., 52) account of more general shifts in historiographical discourse during the early eighteenth century.1111.In her account of the early modern development of the concept of the fact, Shapiro (2000Shapiro, Barbara 2000 A Culture of Fact: England, 1550–1720. Ithaca, NY: Cornell University Press., 52) suggests that in the early eighteenth century, “documentary historians were […] subtly expanding the ‘fact’ from event or deed, that is, actions occurring at a particular moment in time, to cultural conditions, institutions and practices that existed over lengthy periods and were not attributable to individual actors.” For example, in a footnote to one section of Smith’s translation (Example (6)), ‘facts’ are clearly used in a way that equates them with pieces of information necessarily ‘grounded’ in reality (the implication being that facts ‘without any grounds’ are not true facts). These ‘facts’ are ‘established’, not ‘committed’; they are true knowledge about past events rather than the events themselves.

(6)

[FOOTNOTE] The letters of Hippocrates, which mention this affair, are certainly spurious: the facts they would establish are without any grounds, as Le Clerc hath proved to conviction

Similarly, in a passage extracted from his Translator’s Preface (see Example (7)), information regarding “the natural barrenness of the soil” is given by Smith as a specific example of the “certain and indisputable facts” on which Thucydides’ account of the early history of Greece is founded.

(7)

[PREFACE] He begins at the source, and traces the original of the Greek communities from certain and indisputable facts; and the growth of Attica in particular, from the natural barrenness of the soil, which tempted no invasions

‘Fact’ here then does not refer to an action performed by a named agent, but to knowledge regarding an observable and irrefutable feature of the natural world. Nevertheless, it should be noted that this more scientific sense of ‘facts’ is limited in use to just three instances in the main body of Smith’s translation of Thucydides’ Greek, shown in Example (8).

(8)

1. It is a truth evinced by facts, that few considerable armaments of either Grecians or Barbarians, which have been sent out on remote expeditions, have returned successful

2. and the strongest evidence of this he himself hath been pleased to give, by affirming that Ionians have been eternal foes to Dorians. The fact is incontestably true

3. It is a known fact that Aminocles, a ship-carpenter from Corinth, built four ships for the Samians

It is striking to see in these lines that ‘facts’ again are being used as independent sources of proof in support of an argument and/or as synonymous with the objective pieces of evidence showing the truth of a state of affairs. It is also interesting to find in Example (8a) ‘facts’ being presented not merely as actions committed but as agents themselves: in this one instance, facts “evince” the truth and are thus seen to occupy a much more active and decisive function in Thucydides’ History as entities capable of engendering a change in opinion. Yet, as noted above, occurrences of this more modern, scientific usage of ‘fact’ are still relatively few and far between in Smith’s version.

7.3 Bloomfield (1829)Bloomfield, Samuel 1829 The History of Thucydides, Newly Translated into English. London: Longman, Rees, Orme, Brown and Green.

The situation in Bloomfield’s retranslation, written a little under eighty years after Smith’s version, is more mixed. In just one case, ‘fact’ refers to a specific deed or action, as evidenced once again by the dynamic verbs connected to this noun in Example (9) (“caught,” “accomplish”).

(9)

punishing the authors of such machinations, – not only when caught in the fact (for it were hard to catch them), but for what they meditate, but cannot accomplish

In many more instances though, ‘fact’ is used in contexts where it refers to a particular state of affairs observed to be true by the narrator or temporary narrators. This is most commonly as part of the set phrase “in fact,” which appears 14 times throughout Bloomfield’s text, as shown in the two typical concordance lines extracted in Example (10).

(10)

1. and the Athenians sent them aid, under pretext of consanguinity, but in fact from a wish to prevent the introduction of corn thence to Peloponnesus

2. he proceeds to the Hellespont, to cooperate, as he gave out, in the Grecian war, but, in fact, to negotiate matters with the king of Persia

In these lines, ‘fact’ is used as a means of highlighting the contrast between the pretexts and explanations provided by the various protagonists of the History in relation to their actions, and their true intentions, as interpreted by Thucydides. This is subtly different to the use of ‘fact’ and ‘facts’ in Example (8a) and in the concordance lines extracted from the later translations below: in Bloomfield’s translation, ‘fact’ generally refers to a reality maintained to be true by Thucydides; the validity of these statements remains linked to his authority as narrator and to his subjective experience of events.

Interestingly, instances in which ‘facts’ are presented as entities in themselves, used as external proofs of the truth of a statement and having their own independent authority and agency in the text, are once again limited in number, as is the case in Smith’s translation. Indeed, only two examples of this use of the term can be found in the concordance generated from Bloomfield’s version, shown in Example (11). The distinguishing feature of these lines is the presence of reporting verbs in passive constructions (“is afforded by,” “are shown by”) which clearly establish ‘facts’ as objective and indisputable sources of truth, used in support of an argument.

(11)

1. Again, what seems to me a convincing proof of the feeble power of the antients is afforded by the fact that before the Trojan war Greece appears to have achieved no enterprise in common

2. and even these, though the most memorable of all preceding ones, yet are shown by facts to have been inferior to their fame, and to the current report even now prevalent concerning them through the poets

7.4 Dale (1848)Dale, Henry 1848 The History of the Peloponnesian War by Thucydides: A New and Literal Version. London: George Bell and Sons.

Reporting verbs are not much more common in the concordance lines generated from Dale’s translation: only four instances of this patterning can be found in this mid-nineteenth century version of the text, shown in Example (12).

(12)

1. but through want of money both the undertakings before this were weak, and this itself, though more famous than the former, is shown by facts to have been inferior to its fame, and to the present report of it, which has prevailed by means of the poets

2. For let him regard it as it is characterized by the facts of the case, and he will find that there is no reason for its being deemed a state of peace

3. thus one of the allies of the Athenians some time after asked one of the prisoners from the island, by way of insult, if those of them who had fallen were honourable and brave men? to which he answered, that the atractus (meaning the arrow) would be worth a great deal, if it knew the brave men from the rest; thus stating the fact, that any one was killed who came in the way of the stones and arrows

4. and the certain fact respecting those who had done the deed no one was either able to state then, or has since been able

Nevertheless, an intriguing pattern of collocation with verbs of persuasion can be observed, marking an important shift with respect to previous translations. In Dale’s version, ‘facts’ are said to “induce,” “encourage” and “compel” various protagonists in choosing their next course of action, as in the concordance lines in Example (13).

(13)

1. and from this very fact we were compelled at first to advance it to its present height

2. And be not induced by the fact that it is a great naval alliance that they offer you

3. When the light-armed from Spartolus saw these, being encouraged by the accession to their force, and by the fact that they were not worsted before, in conjunction with the Chalcidian horse and the late reinforcement they attacked the Athenians again

These verbs not only have the effect of attributing ‘facts’ considerable agency in the narrative; they also establish ‘facts’ as objective information requiring little or no interpretation on the part of the observer. In this sense, ‘facts’ can be seen almost to speak for themselves.

7.5 Crawley (1874)Crawley, Richard 1874 History of the Peloponnesian War by Thucydides. London: Longmans, Green and Co.

As suggested by the results of the quantitative word frequency analysis in Section 6, it is in Crawley’s translation that we find the strongest emphasis on ‘facts’ as the basis of Thucydides’ History. To an even greater extent than Dale’s version, published 26 years earlier, Crawley has Thucydides deploy ‘the facts’ as authoritative sources of proof in support of a wide variety of arguments across the full length of his work: 35 examples of this use of ‘fact(s)’ can be found in total. Moreover, as can be seen in the concordance lines presented in Example (14), ‘facts’ are in many of these instances positioned not only as the agent of the verb but also as its subject in a series of constructions deploying the active voice: in this translation, facts “point to” a particular conclusion, “show” the truth of a matter and “make it […] incumbent” on certain human characters to act in a particular way. Crawley is the only one of the six translators to adopt this discourse; he would thus seem to be the translator who most clearly foregrounds the idea that the meaning of a fact is given and that it requires no form of subjective interpretation. Indeed, these ‘facts’ can repeatedly be seen to make their own argument.

(14)

1. this is shown by the fact that the temples of the other deities, besides that of Athene, are in the citadel

2. These islanders were Carians and Phoenicians, by whom most of the islands were colonised, as was proved by the following fact

3. the strength of his navy is shown by the fact that his own was the largest contingent

4. Any other view of the case is condemned by the facts

5. And the fact that the people in these parts of Hellas are still living in the old way points to a time when the same mode of life

6. but the facts show that they were Iberians, driven by the Ligurians from the river Sicanus in Iberia

7. Our revolt, however, has taken place prematurely and without preparation – a fact which makes it all the more incumbent on you to receive us into alliance

8. For kindness opportunely shown has a greater power of removing old grievances than the facts of the case may warrant

This idea of the absence of any need for human interpretation in relation to facts is further in evidence through the way in which Crawley presents facts as having an almost material presence as a source of physical energy, most notably through allusion to the metaphor of facts as light (Example (15a–c)). Example (15d) would even suggest that at the mere “touch” of fact, the hyperbole and exaggeration of poetic accounts of the past simply “melt”.

(15)

1. Looked at by the light of facts it cannot, it will be found, be rationally considered a state of peace

2. To this we answer that if any of you imagine that the war is far off, he is grievously mistaken, and is blind to the fact that Lacedaemon regards you with jealously

3. you might be unacquainted with the facts and it would be our duty to enlighten you

4. whose verse might charm for the moment only for the impression which they gave to melt at the touch of fact

7.6 Jowett (1881)Jowett, Benjamin 1881 Thucydides, Translated into English. 2 vols. Oxford: Clarendon Press.

Comparison with the translation produced by Crawley’s near-contemporary, Benjamin Jowett, helps reveal the extent to which the former translator’s presentation of facts may be the result of both a general shift in the norms of historiographical discourse and a specific intervention on the part of this individual translator (see Brownlie 2006Brownlie, Siobhan 2006 “Narrative Theory and Retranslation Theory.” Across Languages and Cultures 7 (2): 145–170. , 156). While the concordance lines in Example (16) show that we can find ‘facts’ being used in support of the historian’s argument in Jowett’s rendering of Thucydides, this patterning is by no means as evident as it was in Crawley’s version, occurring just three times in total.

(16)

1. And traditions which had often been current before, but rarely verified by fact

2. This is proved by the fact that when the Athenians purified Delos

3. And a striking confirmation of my argument is afforded by the fact that Attica through immigration increased in population more than any other region

‘Facts’ are still repeatedly presented by Jowett as things that are known to be true, but this translator does not have Thucydides call upon their objectivity and authority with the same frequency as Crawley. Rather, my analysis would suggest that Crawley’s version stands apart from those of his predecessors and immediate rivals in the manner in which it appears to align Thucydides’ historiographical method with developments in his contemporary Britain concerning the reframing of history as a science.

8.Conclusions

This conclusion is important for the way it highlights the significance of the mediation of translators in shaping the modern reception of ancient historians. It shows that by adopting the modern scientific discourse of facts in his target-language rendering of Thucydides’ text, Crawley provided an interpretation which seems to present the author as much more concerned with objectivity and empirical methods than that of any of his predecessors and even his more contemporary rivals. This may or may not have been a conscious strategy on the part of this late nineteenth-century translator: the historical distance and paucity of extra-textual, biographical details available make any attempt at developing a firm hypothesis in this regard fraught with difficulty.1212.A collection of Crawley’s personal letters and poems was published after his death (see Crawley 1900Crawley, Richard 1900 In Memoriam: Letters and Poems of Richard Crawley. N.p.: Printed for private circulation.). These documents shed light on the translator’s political outlook, as well as on the social circles in which he moved, but unfortunately reveal very little of relevance to the present study. My analysis would imply, however, that the six Thucydidean translators studied here were influenced to varying extents by shifts in attitudes towards the proper aims and methods of history writing over time.

This investigation has of course only scratched the surface in its exploration of the translation of Thucydides at different points in history and more research in this vein is needed. Most notably, comparison with English translations produced during the same period of the works of other classical historians – such as Herodotus, Livy, Tacitus and Plutarch – would usefully help build a picture of the extent to which Thucydides has been translated differently to his ancient peers, or whether all classical histories were re-presented in these ways during the nineteenth century. Indeed, the contrast with translations of Herodotus would be particularly interesting, given the ways in which this ancient author was so frequently framed during the Victorian era as the ‘anti-Thucydides’, the antithesis of Thucydides’ critical approach (Morley 2016Morley, Neville 2016 “The Anti-Thucydides: Herodotus and the Development of Modern Historiography.” In Brill’s Companion to the Reception of Herodotus in Antiquity and Beyond, edited by Jessica Priestley and Vasiliki Zali, 143–166. Leiden: Brill. ). Through such studies, it is hoped that greater interdisciplinary interest may be garnered from across the (digital) humanities in the contribution of translation to the history of ideas and in the active role translators often play in producing and disseminating knowledge within their own societies, as well as on a global scale (Baker 2014Baker, Mona 2014 “Translation as Renarration.” In Translation: A Multidisciplinary Approach, edited by Juliane House, 158–177. Basingstoke: Palgrave Macmillan. ).

Finally, it is hoped that the digital corpus-based methodology developed here will help inspire similar analyses of further terms of relevance to modern scientific discourse within (translated) historiographical texts (such as ‘evidence’, ‘proof’, ‘method’). All data analysis tools and methodologies have their own conceptual blind spots and – as Moretti and Sobchuk (2019)Moretti, Franco, and Oleg Sobchuk 2019 “Hidden in Plain Sight: Data Visualisation in the Humanities.” New Left Review 118: 86–115. have recently argued – computer-assisted text-processing technologies are certainly no exception. Nevertheless, it is suggested that, in combination with close textual analysis of specific passages (e.g., Greenwood 2015Greenwood, Emily 2015 “On Translating Thucydides.” In A Handbook to the Reception of Thucydides, edited by Christine Lee and Neville Morley, 91–122. Chichester: Wiley.), corpus-based approaches – mixing both quantitative and qualitative interrogations of the data, and exploring patterns in language use present across whole texts or collections of texts – might fruitfully open up novel perspectives on the documents, people, practices and cultures we aim to study.