Thick Description beyond the Digital Space

The arrival and rapid spread of digital media can help the humanities better discern general patterns in human societies and cultures, and single out instances or moments which need an in-depth analysis. However, the process of digitization also poses certain threats to the humanities: it introduces new and possibly distorting filters between ourselves and our object of research, it can make us blind to the social, political and cultural situatedness of the sources, and it can cause us to forget that certain aspects of human life are simply too complex to be digitally understood. Briefly elaborating on the case of parliamentary irony in Belgian history, I try to demonstrate the need to combine “flat” digital searches with multilayered hermeneutic approaches based on traditional philological research.


Introduction
The owl of Minerva flies out at night and perceives the essence of things while ordinary citizens are asleep.Those who subscribe to this Hegelian and idealistic representation of philosophy tend to forget that owls are birds of prey that tear apart and eat mice or other small animals.Precisely with this materialistic slant added, however, the metaphor seems appropriate to describe what scholars in the humanities do (although, admittedly, the suggestion of violence it carries does not tie in with the generally peaceful nature of their work and habitus).Like the birds of prey, they try to fly high in order to see general patterns in a world that is populated by human beings-patterns that are not viewed by the people on the ground.In order to actually catch their prey, however, owls, eagles, buzzards, hawks and the like need to dive to the ground and enter its concrete habitat.Most of them do so in full daylight, when most living beings are equally conscious and alert.In the course of that dive, their perspective changes from two-dimensional to three-dimensional: instead of hovering above a flat surface, they are now surrounded by trees, bushes and other elements of the landscape.They enter the "real" world and experience its resistances and complexities.Likewise, scholars in the humanities tend to select concrete cases in order to study them in detail, and most often from different angles.In the case of both the bird of prey-let us call it an eagle from now on-and the humanities scholar, the two-dimensional view from above and the three-dimensional view on the ground mutually reinforce one another.If the bird of prey would never dive to the ground, it would lose the energy it needs to continue its majestic flight; however, in order to localize its prey, it needs to take sufficient height.If the humanities scholar never looks at his or her subject matter from a distance, he or she will fail to discover meaningful cases; however, if he or she never studies cases in depth, his or her view of the global patterns risks remaining sterile.
Unlike the oft-used opposition between the bird's eye view and the frog's perspective, the image of the eagle of humanities implies the unity of the scholarly subject.Instead of an opposition between two beings with totally different interests and world views, it presupposes a fruitful tension between two types of activities carried out by the same person (or by a collective of like-minded persons).
Precisely this tension has always been both one of the main challenges to the humanities and one of its principal charms.The recent digitization of both sources and research methods, however, has given an entirely new dimension to this tension.In the first place, it has dramatically enhanced the possibilities of the view from the sky.It seems as if it has helped to take away the fog which had hitherto prevented the eagle from sharply discerning patterns in the landscape underneath.Until the recent past-and in many fields up until today-scholars in the humanities tried to reconstruct these general patterns by assembling myriads of pieces they had gathered through laborious explorations of the sources, and many corpora of sources needed the combined effort of more than one lifetime to be disclosed in a more or less generalizing way.Today, some of these corpora simply reveal their secrets as soon as someone clicks on the right buttons on his or her computer.The emerging trends can be attractively visualized by means of charts, graphs, digital maps, word clouds or other tools.
If, for example, the French historian Antoine Prost had had available a digital version of the late 19th century electoral proclamations in the early 1970s, he could have gained even more complete, sophisticated and precise results in a very short time-span than the ones he reached in reality after long, laborious and partly collective research.Beyond situating the central words used by respectively right-wing and left-wing politicians on an X-and Y-axis according to their relative frequencies, he could have quantified their co-occurrences, their grammatical position in sentences, and so much more.Moreover, he could have related the words not merely to the political ideology of their authors, but also to their age, region and social or professional profile.With much less sweat, he would have paid a more important contribution to our understanding of political life under the Third Republic [1].
As I argued elsewhere, the digital tools are not only helpful for these macroscopic approaches, but can equally improve the detailed analysis of concrete cases [2].By chasing the fog away, the digital tools not only provide a better view of the landscape in its totality, but equally make it easier for our eagle of humanities neatly to localize the most rewarding animals of prey.Hence, the dive towards this prey can be more secure and can yield an even more satisfactory result.Indeed, the general patterns which emerge from intelligent digital searches immediately lay bare the instances or moments which beg for in-depth analysis.These can be central (or on the contrary marginal) words in a word cloud, peaks or extreme lows on charts, points from where a line starts or ends or where two lines cross one another.Delving into the sources exactly from these points of entry can make us the witnesses of the birth or the end of an evolution, of moments of crisis, or of turning points, be it in general historical evolutions or in the work of one author.Moreover, recognizing the situatedness of these instances in broader patterns adds a layer to their context and makes a deeper understanding possible.It can help us to reach what I formerly called "a higher form of hermeneutics".However, we should not be tempted into an unbridled optimism.While chasing away the fog, the digital tools may open up sharp and attractive views, but these are not absolutely truthful.The fog is in many ways replaced by other filters than those generated by the incomplete conservation of the sources and the limited capacities or the subjective prejudices of the scholarly observer.Those filters can misguide the eagle and cause it to catch only meager pieces of prey.Even more harmful is the potential danger that the eagle is so much enchanted by the view of the landscape that it forgets even to look for its potential prey in it.It neglects to dive, loses its energy, and finally falls on the ground.Both the majestic flight and the pitiful fall belong to the possible fates of the humanities in the digital age.
In this contribution, I will deal with the potential threats of digitalization for the humanities-more precisely with the distortions and the anachronisms it can entail, as well as with its potential to make the humanities less humane.The examples I use to illustrate them will mainly be taken from modern political history-my own discipline-and with language-oriented methodologies, but I believe other disciplines are faced with familiar problems.The example I will elaborate on in the last section of this contribution deals with the history of parliamentary humor and is meant to show that the threats of digitization should not be avoided by escaping from digital tools and methods altogether, but on the contrary by combining them with old-fashioned hermeneutic approaches.Together, they can form signposts to regions which have hitherto been underexplored by the humanities.

To Be Digitized or Not to Be
In the knowledge regime of which each endeavor in the humanities consists, digitization creates one more level, which has to be situated between the actual sources themselves and the researcher.Needless to say that this space was not empty before the arrival of digitized sources.It was inhabited by, among others, archivists, librarians and other institutional actors, but also by relatives of the creators of the source and by all kinds of contingent factors.Together, these elements decide whether or not a source is at all preserved for posterity.The printing press and the ensuing editorial and-especially since the 19th century-academic choices added to this selective conservation of the sources an uneven accessibility.A relatively minor set of originally unpublished sources were now multiplied in varying degrees, dramatically lowering the threshold for consulting them.The decision to publish these sources was, however, motivated to a large degree by contemporary concerns, and therefore introduced an additional element of presentism in the practice of historically oriented scholars.
This same tendency is replicated in an even more extreme form by the arrival of digital media.On the one hand, they are enormously democratizing the historical practice by rendering ever more sources available to an expanding group of users.On the other hand, the question of whether or not they are digitized becomes ever more determining for sources to be included at all in historical research.The answer to that question, however, does not necessarily follow an historical logic.Converting paper sources into digital sources requires technological means and manpower-and, therefore, money.Historians such as Philippe Noiret, Klaus Kiran Patel and Pieter Lagrou have already rightly warned that the political willingness to provide these funds is more likely to be found when an issue with a certain urgency in the present is at stake, or when certain stakeholders have an interest in disposing of these digital sources [3][4][5].That is why in Belgium, during this period of centenary commemorations, an extreme amount of sources regarding the First World War are being digitized, whereas other periods tend to remain more in the shadow.As far as the Second World War is concerned, the Belgian resistance press has been fully digitized [6], whereas the collaborationist or neutral press is still largely to be consulted in paper or microfilm form.As such, a medium that was ephemeral, fragmented and poorly read at the time of its creation has received a more stable and accessible afterlife than the papers that were regularly spread and often massively read during the war itself.
The presentist politics of digitization, however, can be inspired as much by an academic as by a societal logic.Academic disciplines know their fashions that in themselves are often, but not always and seldom exclusively, determined by the demands of society.Sources in which the answers to questions asked in these fashionable research trends can be expected have more chances to get digitized than others.However, even then, much can depend on serendipitous factors such as the specific research context, the juridical framework and/or the digital awareness of the scholars involved.The example of "Ottoman Diplomats" [7], a digital tool in the development of which I have been involved, may suffice to illustrate that point.The database was the outcome of a research seminar with undergraduate students that I coordinated together with the Antwerp doctoral student Houssine Alloul.Undeniably, the very fact that a research seminar was organized around this topic can be situated within the current high tide of Ottoman studies, which in itself ties in with a more general reappraisal among scholars of the Ottoman era as a period yet untainted by the current waves of both secular state nationalism and religious fundamentalism in the Middle East (a tendency which is radically opposed to the imperial neo-Ottomanism cherished by President Erdogan).However, the precise content of both the seminar and the database is as much determined by more contingent factors as by this general political context.For one thing, the database solely contains transcriptions of the reports sent by the Ottoman diplomats in Belgium to their ministry of Foreign Affairs, whereas Alloul's own PhD dissertation deals with the activities of Belgian diplomats in Constantinople.The seminar, then, resulted from his original aim to investigate both sides of the diplomatic relations between the two countries, and from the exploratory research he had done in the State Archives of Istanbul with that objective in mind.For practical reasons, the result of this exploratory research was not integrated in the final PhD project, but the consulted sources were more appropriate for collective research with students and for digitization than the sources extracted from the Belgian archives.This in turn was due to the fact that the archives of the Turkish Ministry of Foreign Affairs were at that time more accessible and less surrounded by secrecy than their Belgian counterparts.The complete series of reports sent from Brussels to Constantinople could not only be consulted freely, but it was also possible to take photographic reproductions of them.In order to get the collection thus gathered transformed into an accessible database, some more elements were needed: the most prominent among these were probably a doctoral student with a high level of digital literacy and a research group with some experience in creating online databases.
The varying states of digitization of different sources also reflect, in other words, diverging digital cultures.These can be situated at the level of individual scholars, of research centers and (sub)disciplines, and also of entire countries.The digitization of 19th and 20th century parliamentary proceedings is illuminative for this last point.Parliaments as institutions were remarkably similar throughout the Western world, and, as a consequence, generated also strikingly similar reports.The process of digitization, on the contrary, has created important divergences with regard to their accessibility and their researchability.When flying over France, for example, the eagle's view of parliamentary history will be seriously impeded by a thick fog created by the non-existence of digital proceedings in any other than a Portable Document Format (PDF) [8].Moving further to the north, this fog will make place for a cloudy sky with bright patches.Only a fragment of the Belgian Parliamentary proceedings-those of the plenary sessions of the House of Representatives-are digitized in a searchable way.Moreover, this database [9] has come about with a very low budget in the context of an academic project at the University of Antwerp.Therefore, it is rather imperfect and unstable, although it does contain interesting tools for large-scale research.When the eagle reaches The Netherlands, the sky has become entirely free of clouds and the view has become nearly impeccable.The proceedings of the Dutch Parliament (both the First and the Second Chamber, both the plenary sessions and the commissions) are entirely digitized as the result of a state-sponsored project carried out by the Dutch National Library [10].Since the quality of Optical Character Recognition (OCR) is nearly perfect, searches in the database yield highly trustworthy results and metadata can be added to it relatively easily.A similar situation can be found in countries like Great Britain, Germany and Norway, whereas the digitization of the parliamentary proceedings in, for example, Sweden and Finland is only in its starting phase.These different paces reflect, to a certain degree, differences in administrative and political culture, but are equally caused by contingent factors.In any case, the current-day state of digitization of parliamentary proceedings does not reflect the relative importance of the national parliaments in European or world history.The consequences of this state of affairs are anything but minor.In the comparative histories of parliamentary discourse and culture which start to be produced, a country such as France occupies, at best, a marginal place solely because of the poor state of its digital proceedings [11,12].A comparison with, for example, Canada turns out to be more feasible than one with France, even if in the 19th century the impact and notoriety of the latter's parliamentary system was much more important.Ongoing projects of the homogenization of digital proceedings, such as dilipad (Digging into Linked Parliamentary Data [13]), therefore have to work with rather random selections of countries.
Even among those countries with a well-advanced state of digitization, however, transnational comparisons can be less easy than they seem at first.This can be due to the minor or major differences between the original sources or the languages in which they are written, but the difficulties can equally arise from the choices which have been made in the digitization process itself.Again, the parliamentary procedures can illustrate this point well.In most cases, the option has been simply to apply Optical Character Recognition to photographic versions of the sources.This makes it possible to search for specific (combinations of) terms throughout an immense corpus of text, but the precise legislative context in which they occurred has to be reconstructed through a careful reading of the passages involved.The parliamentary services of Norway, on the contrary, have opted to index the documents from the start, and to group them according to the legislative process to which they belonged.Search engines can be applied to the indexes created thus, and not to the text corpus itself.This choice makes their database extremely useful for legal historians, but much less for those who are interested in the evolution of parliamentary discourses and cultures.As long as no international organization will steer digitization programs, they will tend to instill a new complexity into endeavors of international comparison [14].
However, even if such an instance would exist and would succeed in developing internationally recognized standards for digitization procedures-even then it is far from certain that the arising digital corpora would be a perfect reflection of the source situation as it existed before the pre-digital age.Indeed, other contingent factors can equally contribute to a distorted image.One of these factors is the rules regarding copyright.Because an organization like the Belgian Royal Library wants to abide by these rules carefully, it can only make a relatively small portion of its digitized newspapers actually accessible online [15].Strict copyright clauses apply particularly to newspapers whose importance and continuity is attested by the fact that they exist today.Consequently, a sort of inverted presentism can arise: documents whose current-day relevance seems doubtful have a greater chance to be rendered accessible online.That is also the case for a database like the Digitale Bibliotheek voor de Nederlandse Letteren [16], which aspires to put online as much Dutch literature as possible.Although it tends to prioritize canonical literature, some of the major works and journals remain absent from the site because of copyright issues.

Humanities beyond the Limits of the Digital World
Intrinsically, none of these objections really forms a fundamental problem.It is self-evident that the humanities scholar should always evaluate the usefulness of his or her sources by assessing not only their origin, but also their dissemination and transmission.Digitization seems to add only one more layer to their transmission history.However, whereas former stages in the evolution of the communication media tended to raise the philological awareness, digitization on the contrary threatens dramatically to lower it.Indeed, generations of scholars in the humanities have spent the best of their lives comparing different manuscript versions of one text, or these manuscripts with the published version, the published version with its translations to other languages or their movie adaptations, and so on.On the other hand, they tried to unravel the true logic of an event by delving deep into several archives, often without the certainty of finding anything at all.Doing this work did not only involve perspicacious interpretive acts, but also required the passing of several practical thresholds (traveling, keeping in mind opening hours of libraries and archives).If these resistances mortgaged the possibilities of large-scale research, they helped to develop a keen sense of the embeddedness of sources and genres.If one wanted to follow the trajectory of a certain intellectual current or political movement in the past, one first had to build up a thorough knowledge of its press, its institutional set-up and its central personalities before looking for the places where it could possibly have left traces.The actual search for these traces had to be based upon a well-grounded intuition where the most relevant material was to be found.This heuristic process itself formed part and parcel of the eventual interpretation of the evidence and of the construction of an intellectual argument.
Digitization, on the contrary, makes it possible to cut bluntly and without effort through source types and genres in search of patterns of behavior or significance.In the digital era, for example, it is extremely easy to find references to even lesser known historical persons in the most diverse and often unexpected sources: he or she can make his or her appearance in a small announcement in a regional newspaper or in the digitized inventory of the personal papers of someone who was only obliquely related to him or her (and thus in places where before we would never have bothered searching).Furthermore, we can trace the fluctuations in the use of a concept throughout millions of books and articles at one time.Needless to say that this state of affairs admits us to accumulate our knowledge at a pace which was hitherto unthinkable.Rejecting these potential advantages would therefore be simply absurd.Still, precisely the easiness with which we can gain knowledge in the present day circumstances can make us forget to ask the necessary questions about the embeddedness of the sources in their social, cultural and political contexts, or to situate them within a specific genre and the conventions it entails.Moreover, it can strip us from the specific linguistic and contextual knowledge we need in order to answer these questions, or even from the willingness to build up this knowledge slowly and carefully.Many of those who work on a daily basis with digital natives as history students will no doubt recognize that this threat is far from imaginary.When asked to give an overview of the sources they will use in order to answer a historical question, it is not unusual nowadays for students to simply answer with the name of the digital database in which they will be able to find these sources.The original sources lose their historical specificity behind the digital veil, and the students are lured into intellectual laziness and anachronistic thinking.
Beyond this heuristic trap, digitization can also make us unaware that a part of human behavior is so complex and multilayered that it can never be understood by digital means, however sophisticated they may be.When that happens, the eagle of humanities will have grown lazy and weak.Its view will be as flat as it is wide.It will have forgotten the sheer existence of the three-dimensional view, but also the taste, the smell and the warmth of its prey.
Obviously, this is not a plea to abdicate from our attempts to make the digital tools ever more performant, and to let them guide us ever more in our attempts to understand the workings of human societies.On the contrary even, students in the humanities should be initiated more than they are today to the possibilities of the digital humanities, but no less should they be reminded of their limits, and of the immense interpretive potential situated beyond these limits.They should get accustomed to the resistance historical sources offer to interpretive acts, which will therefore always remain imperfect.However, they should equally receive the equipment they need to carry out these acts nonetheless, and to appreciate the beauty of the imperfect results they gain by carrying them out.They should learn, in other words, how to combine the delights of a wide but flat view with the pleasures of thick description.

Safe for the Digital Hunt: Irony in the House
In order to illustrate this point, let me return one last time to parliamentary history, and more precisely to an aspect which, during the last decades, has gained some scholarly attention: parliamentary humor.This interest in parliamentary humor is not merely a gimmick of postmodern scholars keen on importing insights from historical anthropology.Humor is a potent tool for creating collective identities through the combined working of exclusion (laughing at) and inclusion (laughing with).In a political arena, where it is important to forge friendships and reinforce antagonisms, the importance of such a tool is undeniable.Scholars such as Antoine De Baecque, Joseph Meisel and me, therefore, have tried to find out which side of the political spectrum was most active in using this tool and whether various ideologies converge with different varieties of humor [17][18][19].Invariably, however, they were confronted with the extreme difficulties raised by these questions: because of its omnipresence, systematically tracing all the occurrences of humor in the immense corpus of the parliamentary proceedings turned out to be impossible.Hence, these studies are inevitably limited to micro-historical approaches or construed around debatable conjectures.
The digitization of the parliamentary proceedings at first sight offers an incredible tool to overcome these difficulties, especially in those countries where the parliamentary services have decided from the start to insert para-textual elements such as "laughter" and "hilarity" in the proceedings.These insertions are often very precise, and indicate whether the laughter came from the right, from the left or "from all benches".In some cases, it was specified whether the laughter was "ironical" and even "smiles" did not go unnoticed.In the Belgian and the French cases, for example, formulas such as "ironic smiles from the right" or "ironic laughter from the left" are relatively frequent, and can be traced by simply typing them in the advanced search module of the digital database containing the proceedings of the Belgian Chamber of Representatives [9] (with the renewed caveat that due to the imperfect Optical Character Recognition, we cannot be sure that all occurrences have been tracked).Doing so, it seems tempting to make some general statements about the history of irony in the Belgian chamber of representatives.As such, it appears that the first mention of "ironic laughter" was made in 1857, but that it would occur no less than 2594 times between that moment and 1999.It knew its heydays between 1879 and 1913, when it occurred 1377 times, which means an average of nearly 70 times a year.After the First World War, this figure would fall back to only a bit more than 20 times a year.The end of the Second World War seems to have announced a new flourishing of the ironic laughter, which was recorded 556 times between 1944 and 1961 (or an average of 30 per year).After that moment, it nearly entirely disappeared from the proceedings.The "ironic smiles" appeared much more sporadically (only 40 times between 1901 and 1971), but their distribution over time followed a more regular pattern, with only a remarkable "boom" between 1901 and 1905.
If we break these figures down further, we can observe that "ironic laughter from the right" occurred 803 times, "ironic laughter from the left" only 501 times.However, to the latter should also be added the 114 occurrences of "ironic laughter from the socialist benches" and the 191 occurrences of "ironic laughter from the extreme left" (which, until 1914, equally referred to the socialists), as well as the 27 pre-1914 occurrences of the "ironic laughter at the benches of the opposition" (which in that period was formed by a left-wing alliance of Socialists and Liberals).Counted together, this would mean that "the Left" tended slightly more toward ironic laughter than the right, certainly if we would recalculate it pro capita.The fact that the pattern of the "ironic smiles" is largely similar could be invoked to corroborate this assertion.However, the figures also show that the "communities of irony" became less stable after the First World War.The standard references to "the left" and "the right" were gradually replaced by new groups such as the "Flemish Nationalists" or the "Rexists" (an antiparliamentary right-wing party of Belgian Nationalists) by linguistically limited versions of older groups, such as "the Flemish Catholics", or by new combinations, such as "the Liberals and certain benches on the right".After the Second World War, references to ironic laughter and smiles would most often be connected with specific parties (e.g., "ironic laughter on the benches of the Parti Social Chrétien").
There is no doubt that these data tell a lot about shifting identifications in the Belgian Chamber of Representatives, where a stable divergence between a confessional Right and an anti-confessional Left was gradually complicated by crossing cleavages based on class, language and devotion to parliamentary politics.They would do so even more if we would dispose of exhaustive metadata, which would enable us to see to which parties those whose interventions provoked ironic laughter or smiles belonged.However, do these results make us any wiser about the history of irony in the Belgian parliament?Can we conclude that irony only appeared in the House in the 1850s and waned during the interwar period?Can we assert that the right had a slightly more pronounced talent for irony than the left?Obviously, the answer to these questions can only be negative.Even if we would agree upon a workable definition of irony-as, for example, any rhetorical or performative device creating a deliberate opposition between what is said or done on the one hand and what is meant on the other-it is very hard to know whether the term was used in the same sense by the parliamentary stenographer who decided whether a laughter or smile was ironical.We do not know which criteria he used to do so.Are we sure that he noticed all smiles in parliament, let alone all ironic smiles?Even if, miraculously, that were the case, though, the results would only tell us something about a small part of the ironical behavior in the Chamber.Laughter and smiles can be the external manifestation of an ironic interpretation of parliamentary interventions.Digital tools can impossibly teach us why precisely these interventions provoked ironic reactions.We can safely hypothesize that in most cases, these interventions themselves were not at all meant to be ironical.On the contrary even, ironic laughter is often meant to delegitimize statements made in full earnestness.Intentional irony, in its turn, does often not provoke any laughter.If we want to study parliamentary irony seriously, we should move far beyond the mapping of reported ironical laughter and smiles.On the one hand, we should read the interventions which provoked them, and interpret them in their short-and long-term context.On the other hand, we should keep an eye on interventions which were intended to be ironical, but which were not followed by laughter.In order to do so, we cannot escape from the micro-historical and hermeneutic approach.A randomly chosen passage can illustrate this.It is taken from a debate about the use, in a state-subsidized school, of two school books in which the Spanish Inquisition was justified and the Italian unification was rejected as a form of banditry.Jules de Trooz, the Catholic minister of the Interior and of Education, answered by asserting that the liberty of education was a sacred principle inscribed in the Belgian Constitution of 1831.He also added: That same language has been held by De Theux, by De Decker, by Malou, by Beernaert, that is to say by all those who have had the great honor to direct our country's affairs in the name of our party.But, if at the Right we have always professed the most absolute respect for the Constitution and for the it proclaims, we have also considered it always a sacred right to enjoy these liberties.In Belgium, there is no State religion, we are gifted with the liberty of conscience and the liberty of cult [20].
At this part of De Trooz's speech, the parliamentary reporter noticed "ironical smiles at the extreme left".The Antwerp socialist Member of Parliament (MP) Modeste Terwagne exclaimed, "How impudent!"(Quel toupet!), which in itself was met with (further unspecified) "Hilarity".De Trooz reacted calmly to this exclamation by saying: "When you will be recovered from your emotions, dear colleague, you will ask the floor, and you will prove the contrary!" The only part of this short passage which betrays a clear intention to be ironical does not precede but follows upon the ironical smiles.Obviously, De Trooz's concern for Terwagne's emotions was not real, but formed part and parcel of a political game.After his opponents from the extreme left had tried to delegitimize his words with their ironical smiles and with an outburst of indignation, he in turn tried to delegitimize their attack upon him by presenting it as the expression of a mental shock.By psychologizing it, the attack upon him was de-politicized.This ironical strategy did not provoke laughter, but it did seem to be successful in silencing De Trooz's opponents-at least for a little while.De Trooz's words which preceded the Socialists' ironical smiles had been everything but ironical themselves.They contained a solemn profession of belief in the Belgian Constitution, and summed up a pantheon of Catholic leaders in which De Trooz tried to situate himself.The ironic smiles, therefore, expressed without words a different vision of Belgian political history-one in which the Catholics were not the true defenders of the Belgian Constitution, but those who abused that same Constitution in order to regain the ancient position of the Catholic world in Belgian society.Parliamentary irony, therefore, can merely be understood if we are familiar not only with the "culture wars" between the opposing forces in Belgian politics and with the central place of education in them, but also with the antagonistic collective memories they brought into the parliament.Just like irony, however, memory is one of those analytical concepts which cannot be captured quantitatively or digitally.We can only understand it by fully and thickly reading the sources, trying extract ever more historical meaning from them by setting them in their diverse and interrelated contexts-much like Clifford Geertz famously did with the wink in the classroom or the practices of the Balinese cock-fighters ( [21], pp.2-30, 412-53).In the specific case, such a "thick description" would ideally entail an investigation into the opinions formulated elsewhere by De Trooz on De Theux, De Decker, Malou and Beernaert, and into their interpretations of the Belgian Constitution.Only then will we be able to fathom precisely why De Trooz's words were considered untruthful by his opponents and therefore became the object of ironic laughter.
In order to find these references, digital tools will doubtless be very helpful, but they can never suffice to reach these multilayered interpretations.These need micro-historical forms of thick description, informed by thorough contextual and biographical knowledge.Hence, it will always remain necessary that scholars undertake this kind of old-fashioned research, which nowadays risks being marginalized by the logic of academia (both at the level of the decision-makers and on the floor).In order to acquire meaning, moreover, the results of these micro-historical investigations will always need scholars who possess what Paul Veyne has famously called the art of telling history as "a true novel" ( [22], p. 10).Indeed, only narratives can fully do right to the complexity of historical (or literary, or other) events, as knots where various larger evolutions come together with each other, with contingent factors and with the free will of historical actors who carry their own life stories.

It Does Matter
Maybe more than anyone else, scholars in the humanities should be careful with metaphors.Nonetheless, in this contribution I have recurred insistently on the metaphor of the eagle in order to depict an ideal image of the humanities in the past, the present and, above all, the future.In this image, deliberately less distant and irenic than that of the owl of Minerva or the "bird's eye view", the alternation of the majestic flight and the agile dive to the earth symbolizes the fruitful and necessary collaboration between distant and close reading, between macro-and micro-analysis, between flat and thick description.However, could it not also be argued that the prey of hermeneutics and micro-analysis is simply too heavy for the eagle to take off again?Should we not, at a time when the Internet enables us to engage in "big history" projects encompassing 13.8 billion years, throw overboard the petty details of men's lives and works?Therefore, is this plea for hermeneutics not simply a rearguard fight of a nostalgic historian?
I dare to believe it is not.As tempting as it may be to walk through history with seven-league boots and see the evolution of mankind-reduced to its main lines-pass by, it will never help us better to understand humanity.Indeed, humanity consists of human beings, whose life is not enacted in millennia, but in a handful (or two, at most) of decades, and is shaped by memories of recent or less recent pasts, by contingencies, conjunctures and choices in the present, and by vague plans or concrete dreams for a near and sometimes distant future.Far from being the masters of their entire fate, human beings nonetheless have the freedom to craft a life with the contextual elements they face.Both when studying the smallest details and the largest patterns of human behavior and expression, the humanities should never lose sight of this laborious but creative confrontation between men and their contexts.If they did, they risk to surrender to determinism and thus to contribute to a world in which human actions are deemed to make no difference.The huge risks to which future generations are exposed (climatic change, growing inequality, etc.) might simply be accepted as irreversible trends.The humanities' essential interest in the daily struggles, dreams and disillusions of human life, as well as their eagerness to reconstruct them meticulously, is based upon their belief that these human concerns do matter.Hence, the humanities do form and have to remain a bulwark against both utopianism and determinism.