Benedict Anderson：Frameworks of Comparison

In my early days at Cornell, use of the concept of 'comparison' was still somewhat limited. I do not mean that comparisons were never made; they were made all the time, both consciously and (more often) unconsciously, but invariably in a practical way and on a small scale. Even today, in the Cornell College of Arts and Sciences, there is only one department (comparative literature) that uses the term in its title, and this department did not exist in the early 1960s when I left for Indonesia to undertake fieldwork.

Historians, anthropologists, economists and sociologists rarely thought systematically about comparison at all. The political science department was a partial exception, since it had a subsection called 'comparative government', to which I belonged. But the comparisons my classmates and I studied were primarily focused on Western Europe. The reason for this was understandable. European countries had for centuries interacted with one another, learned from one another, and competed with each other. They also believed that they shared a common civilization based on antiquity and different Christianities. Comparisons seemed both simple and relevant.

For me, the odd thing was that comparative government did not cover the US itself, which was the preserve of a different subsection called American government. On one level, this division was easy to understand in practical terms. The undergraduate students, thinking about future careers as politicians, bureaucrats, lawyers and so on, were overwhelmingly interested in courses about their own country's politics. The same 'nationalist' interest can be found in most countries. My department was dominated by Americanists because of student demand. A less obvious factor was the pervasive 'frog under the coconut shell' mentality created by what I call 'official nationalism'. The US has two important neighbours, Mexico and Canada, but there were no courses on these countries' politics, and, right up to the point of my retirement in 2001, it was rare to find a student who could name either the president of the former or the prime minister of the latter.

One of the central myths of American nationalism has long been 'exceptionalism'--the idea that US history, culture and political life are by definition incomparable. The US is not like Europe, not like Latin America, and absolutely not like Asia. Needless to say, this fancy is absurd. In different ways, depending on which countries in what periods are relevant, the US is perfectly comparable, especially with Europe, South America, Japan and the British Dominions of the Empire (Canada, Australia, New Zealand, South Africa and so on). Another aspect of this perspective is its deeply ingrained provincialism. Hence the strong resistance to the logical case for including American politics within comparative politics.

One could plausibly add two other factors that are more specific. The first is the institutional history of the study of politics in the United States. One clear relic of this history is that there are still a number of political science departments that call themselves departments of government (Harvard and Cornell among them). Their lineage derives from the merging of law (mostly 'constitutional' law) and public administration, both eminently concerned with the practicalities of governance. In Europe the lineage was quite different: departments of philosophy, sociology, economics and politics based on the grand tradition of Machiavelli, Smith, Constant, Ricardo, Hegel, Marx, de Tocqueville, Weber and so on. My department had a subsection called Political Theory, which was usually taught by a European scholar. Its range extended from Plato to Marx, but included no Americans.

The second factor is that Americans are a practical and pragmatic people, not naturally given to grand theory. A quick glance across the social sciences and humanities for the 'great theorists' of the past century makes this abundantly clear, whether in philosophy (Wittgenstein, Heidegger, Derrida, Foucault, Habermas, Levinas, etc.), history (Bloch, Braudel, Hobsbawm, Needham, Elliott), sociology (Mosca, Pareto, Weber, Simmel, Mann), anthropology (Mauss, Lévi-Strauss, Dumont, Malinowski, Evans-Pritchard) or literary studies (Bakhtin, Barthes, de Man, etc.). All these foundational figures are European.

The grand American exception is Noam Chomsky, who revolutionized the study of linguistics, and, perhaps to a lesser extent, Milton Friedman in economics, though Keynes may last longer. Of course, this does not mean that contemporary US universities are not obsessed with 'theory', only that the 'theory' either comes from outside America, is modelled on economics (which has a strong theory-orientation important for understanding the functioning of modern society), or is underpinned by America's egalitarianism: 'everyone', so to speak, 'can and should be a theorist', though history shows that individuals genuinely capable of producing original theory are rare. My own experience as a student at Cornell occurred before 'political theory' really took hold. My thesis (1967) could almost have been written in a history department. But by then what was later remembered as the era of 'behaviourism', understood as making the study of politics 'scientific', was on the rise.

The thirty-five years I spent as a Professor of Government at Cornell taught me two interesting lessons about US academia. The first was that 'theory', mirroring the style of late capitalism, has obsolescence built into it, in the manner of high-end commodities. In year X students had to read and more or less revere Theory Y, while sharpening their teeth on passé Theory W. Not too many years later, they were told to sharpen their teeth on passé Theory Y, admire Theory Z, and forget about Theory W. The second lesson was that--with some important exceptions like the work of Barrington Moore, Jr.--the extension of political science to comparative politics tended to proceed, consciously or unconsciously, on the basis of the US example: one measured how far other countries were progressing in approximating America's liberty, respect for law, economic development, democracy, etc. Hence the rapid rise, and equally rapid fall, of an approach that today looks pretty 'dead'--modernization theory.

Needless to say, there was often an openly stated Cold War objective behind this kind of theory. Namely, to prove that Marxism was fundamentally wrong! In its innocence, this kind of 'look at me' theory typically ignored such embarrassing things as the very high rates of murder and divorce in the US, its hugely disproportionate Black prison population, persistent illiteracy and significant levels of political corruption, and so on.

Nonetheless, there is no doubt in my mind that my experience as a graduate student unconsciously prepared me for later comparative work. My duties as a teaching assistant in American politics and (European) comparative politics obliged me to study a great many texts that I would not otherwise have read. The undergraduates in those days were 90 per cent American and knew very little about Europe. To help them, I found it useful to make constant comparisons between the US, the UK, France and Germany. I myself took graduate courses on the Soviet Union, Asia, the US and Western Europe. Finally, the format of the Southeast Asia program forced me not only to start thinking across the region in a comparative sense, but also to read across disciplines, especially anthropology, history and economics. At the time, I did not have a high level of consciousness about all this--it was all fun because it was so new to me.

It was still from within this framework that I wrote my first explicitly comparative work, a long article entitled 'The Idea of Power in Javanese Culture', published in 1972 in a book titled Culture and Politics in Indonesia, edited by Claire Holt. This essay had an unlikely origin. One day, as I was sitting in my office with the door open, two senior professors walked by, chatting loudly on their way to lunch. The man doing most of the talking was Allan Bloom, who much later published a best-seller called The Closing of the American Mind. He was a rather fascinating and even intimidating figure. 

Unashamedly effeminate, and clearly favouring his male over his female students, he was nonetheless a charismatic conservative lecturer, and a firstclass scholar in the field of political theory (Plato to Marx). At the University of Chicago he had been among the top students of Leo Strauss, the famous political émigré from Nazi Germany and a principled philosophical conservative, many of whose pupils (especially bright and ambitious Jews) went on to lead the neo-conservative movement in American political life under Reagan and the two Bushes, as well as in the best universities.

What I overheard Bloom say was this: 'Well, you know that the ancient Greeks, even Plato and Aristotle, had no concept of “power” as we know it today.' This casual, lunch-hour comment seeped into my mind and stayed there. It had never occurred to me that the two philosophical masters, whom we were always told to revere as the founders of 'Western Thought', had no idea of power in their heads. Dubious at first, I rushed to the library to consult a Classical Greek dictionary. I could find 'tyranny', 'democracy', 'aristocracy', 'monarchy', 'city', army', etc., but no entry for any abstract or general concept of 'power'.

When writing the final version of 'The Idea of Power in Javanese Culture'--conceived as a study in comparative political philosophy--I tried to anticipate and forestall the easy reaction of most Western readers: 'Well, the Javanese were and are primitive, and we are not.' Luckily, I had help from Max Weber, who introduced the concept of 'charisma' into modern sociology, though he had great difficulty in explaining it clearly and systematically. Hitler, Reagan, Mao, Evita Peron, de Gaulle, Soekarno, Gandhi, Fidel Castro, Lenin and Khomeini: what rationality lay behind their hold on people's imaginations? Was there a substratum of old ways of thinking about 'power' (mana, tédja) even in cultures that thought of themselves as completely modern? Much later on I was gratified to learn that Reagan never made important decisions before his wife had telephoned her fortune-teller, and that the top leaders of today's Chinese Communist Party eagerly consult astrologers and feng shui masters--out of the limelight, of course.

The two main points I want to make about this article are, first, that I began making comparisons from a nationalist point of view, and within an East versus West framework long popular among Orientalists--but in this comparison I wanted to show that the Javanese or Indonesians can be seen as just as 'rational' as Westerners and other peoples, so long as we understand the basic assumptions of their thinking. Second, that taking this approach occurred by sheer chance: I happened to be both Bloom's junior colleague and Moertono's friend.

For the next ten years, however, I really did nothing seriously comparative, and when I returned systematically to the question of comparisons, my outlook and interests were completely different. Even a brief look at 'The Idea of Power in Javanese Culture' (1972) and Imagined Communities (1983) will immediately reveal how far apart they are. For sure, it was partly a matter of age. In 1972, I was thirty-six years young, still untenured, and recently expelled from Indonesia. In 1983, I was forty-seven years old, a full professor, newly appointed director of Cornell's Southeast Asia Program, and busy with the study of Siam. But the age difference was 

by no means the most important factor. Here I would like to jot down some notes about three powerful influences on me over that decade--in no special order.

I was fortunate to have a more intelligent, slightly younger brother, known to the world as Perry Anderson, but within the family by his original Irish name, Rory. For a long time after I left for America, we did not keep in touch very well, except via my mother and sister. After graduating from Oxford in history, I think in 1959, he plunged into Marxist politics and intellectual life. Along with some of his Oxford friends, he quickly moved to work at the recently established New Left Review, to revive and modernize a leftist politics in the UK that had long become fossilized under the uninspiring aegis of the British Communist Party.

The founders of the NLR were Edward Thompson, the great radical historian of rural and working-class England, and the Caribbean social thinker Stuart Hall, who would later become known as the founder of cultural studies. The 'Young Turks' had only affection and respect for Hall, but relations with Thompson were often difficult. He was a brilliant man, but English to his bones, and in some ways a 'Little Englander' with a traditionalist hostility to the intellectual traditions of Continental Europe. My brother and his friends believed strongly that British intellectual isolation had to be broken out of by 1) a massive importing (in translation) of the works of key Marxists beyond the Channel: Sartre, Merleau-Ponty, Althusser, Debray, Adorno, Benjamin, Habermas, Bobbio and many others; and 2) making the NLR as internationalist as possible in the problems it addressed.

Eventually a furious Thompson left, and the young generation took over. At the same time, Rory was busy working on his gigantic project of reframing the whole of 'Western history', which led to his path-breaking books Passages from Antiquity to Feudalism (1974) and Lineages of the Absolutist State (1974), both of them fundamentally comparative. As a 'good brother', I read these books with awe and pride. They showed an encyclopaedic historical knowledge, a mastery of classical prose, and a formidable capacity to sustain a complex but clear argument across hundreds of pages, hundreds of years and dozens of countries.

From 1974 I started to read the NLR from cover to cover and was profoundly re-educated in the process. Here I came into contact with the work of Walter Benjamin, which had a decisive impact on me, as readers of Imagined Communities will immediately recognize. On visits to London, I began to meet the NLR circle and make friends among them. I liked and respected no one more than Tom Nairn, the Scottish nationalist-Marxist who in 1977 published his polemic The Break-up of Britain, which caused a real uproar and led to a stinging attack from Eric Hobsbawm, then the leading figure among the older generation of Marxist historians.

During this process, my brother and I became close again, as we have remained till this day, and he was my key counsellor in preparing the final version of Imagined Communities. Had I not had a brother like him, I am not sure what would have become of me. Through Rory and his friends at the NLR, I became more internationalist and no longer just an Indonesian nationalist.

The second major influence on me was my Cornell contemporary and close friend James Siegel, who is today, in my opinion, the most arrestingly original anthropologist in the US. He had been one of Clifford Geertz's last students before the famous man, enraged by the rowdy student radicalism of the late '60s, abandoned teaching for an aerie at Princeton's Institute for Advanced Studies, where for a long time he was practically the only social scientist. Jim and I had done fieldwork in Indonesia at the same time, he in Atjeh and I in Java. We first met in Medan, a city in northern Sumatra, in the spring of 1964, and immediately became friends. His thesis, published later as The Rope of God, was unlike any anthropological work previously written on Indonesia, and has always been one of my favourites.

How he came to Cornell is an interesting story in itself. Around 1967, an anthropology post came up at Cornell for a young Southeast Asianist, and Jim, among a number of others, applied. In those radical days, candidates were no longer interviewed only by professors, but also by graduate students. When the final decision had to be made, most of the faculty were in favour of James Peacock, who had written a thesis on ludruk, the popular urban theatre of East Java, which he characteristically called Rites of Modernization, and which was published with the same title. This kind of Parsonian title did not help his cause with the students, for whom 'modernization' was an abandoned fetish. They voted overwhelmingly for Jim, and the faculty gave in.

Jim was and still is one of my best friends. We often taught courses together, including one seminar in which we insisted that every student speak in Indonesian! It was he who seriously introduced me to high-class anthropology, including the inspiring Africanist work of the British Catholic Victor Turner. He also made me read Erich Auerbach's Mimesis, an extraordinary account of the history of 'representation' in the West from Homer to Proust. Our favourite class was a joint seminar on the fiction of Indonesia's great writer Pramoedya Ananta Toer, who was then still in one of Suharto's gulags. Careful, close-up reading of fiction with a group of excellent students was quite new for me. Thanks to Jim, I began to think about how I could use my early training in Classical and Western European, as well as Indonesian, literature for a new kind of analysis of the relations between 'imagination' and 'reality' in the study of politics.

The third influence came from students in the Southeast Asia program. They had little interest in the formal American concept of a giant Southeast Asia zone as such. But they acquired smaller types of solidarity among themselves. Anger at the long grim dictatorships in Buddhist Siam and Burma, Islamic Indonesia, the Catholic Philippines, etc., moved the youngsters to rejection. In English they could exchange information that was heavily suppressed in their home countries. They got used to making new comparisons that they had never before imagined.

As for the comparisons typical of Imagined Communities, they were shaped by the book's polemical intentions. Almost all the important 'theoretical' works written on nationalism after the Second World War were written and published in the UK (Miroslav Hroch's pioneering comparative study of 'small nationalisms' in Central and Eastern Europe, written in German in Communist-governed Prague, had to wait a long time to be translated into English). Almost all were written by Jews, though of widely different political outlooks.

On the far right was Elie Kedourie, who was born and raised in the old Jewish community of Baghdad, moved to London as a young man, and came under the influence of Michael Oakeshott, then Britain's best-known conservative political philosopher. On the moderate right was Anthony Smith, a British-born practising Orthodox Jew, who taught history in London throughout a long career. Convinced that the Jews were the most ancient of nations, he consistently argued that modern nationalism grew out of long-standing ethnic groups. On the liberal left was the philosopher, sociologist and anthropologist Ernest Gellner, a Czech Jew born in Prague, who made his way to London just after the end of the war. A sturdy Enlightenment liberal, he pioneered the so-called constructivist view of nationalism, arguing that it was strictly a product of industrialization and modernity. On the far left was the grand historian Eric Hobsbawm, of partial Jewish descent, born in colonial Egypt and substantially educated in pre-Nazi Austria. Hobsbawm was a constructivist as well as a communist, and made a striking contribution to the growing debate on nationalism in the UK with The Invention of Tradition (1983), a collection he compiled with Terence Ranger. The odd man out was Tom Nairn, strictly Scottish, and a New Left Marxist radical.

All these people lived either in London or in nearby Oxford or Cambridge, and they all, more or less, knew each other. All except Nairn were very attached to the UK, partly because it was largely uncontaminated by fascism and violent anti-Semitism, and partly because the state, including England, Wales, Scotland and Northern Ireland, was felt to be more like supranational (if now defunct) Austro-Hungary than standard European nation-states such as France, Italy and Sweden. All these men were basically Europe-oriented, even if Gellner studied in the Maghreb and learnt some Arabic, while Kedourie wrote a lot about his native Iraq, and obviously knew Iraqi Arabic well.

This then was the wide, but very 'British' circle at which Imagined Communities was aimed. The debate was really triggered by Nairn's polemical The Break-up of Britain, which argued that the UK was a fossilized, conservative and imperialistic relic of the past, doomed to break up into its four constituent underlying nations, with Scotland leading the way.

The book was strongly attacked, especially by Hobsbawm, who declared that no true Marxist could be a nationalist; Marxism had been committed from the start to internationalism. I liked the book very much, for its own sake, but also as an Irishman (Southern Ireland, after centuries of English colonial rule, had only won its independence, by armed struggle, in 1922). I did not think of Imagined Communities as a strictly academic book, and it never occurred to me at the time that it would eventually have a wide international audience.

Many people have complained that Imagined Communities is a difficult book and especially difficult to translate. The accusation is partly true. But a great deal of the difficulty lies not in the realm of ideas, but in its original the book contains so many quotations from, and allusions to, English poetry, essays, histories, legends, etc., that do not have to be explained to English readers, but which are likely to be unfamiliar to others. There are also jokes and sarcasms only the English would find amusing or annoying. For fun I always titled British rulers as if they were ordinary people, e.g. Charles Stuart for Charles I, but used the standard format for foreign kings (Louis XIV). A radical English feminist once wrote to complain about this 'discrimination'.

Of course I was pleased. When in the late 1980s two of my best students, Shiraishi Takashi and Shiraishi Saya, decided to translate the book into Japanese, I reminded them that it was not originally intended for Japanese readers, so that they should feel free to substitute appropriate Japanese quotations, allusions and jokes where they liked. I think they were happy to have this freedom.

Imagined Communities was formed in a wider polemical framework than The Break-up of Britain. The first target was the Eurocentrism I saw in the assumption that nationalism was born in Europe and then spread out in imitated forms to the rest of the world. But it was also plain to me that nationalist movements had their historical origins in North and South America, as well as Haiti, and that these movements could not be explained on any 'ethnic' or linguistic basis.

The second target was traditional Marxism and liberalism. Nairn had rightly argued that this kind of Marxism had largely sidestepped nationalism, and had never been able to explain its vast world-historical power. But he had not really attempted to offer a Marxist solution to the problem. I had become convinced that a solution was possible if one took into account the peculiarity of printed books, which began to be published in large quantities in Europe in the sixteenth century. Books were certainly commodities produced by early capitalism, but they were also containers and purveyors of ideas, emotions and imaginings, unlike beer or sugar. Classical liberalism had the same failings.

The final target was a powerful tradition that treated nationalism as if it were just another 'ism' alongside liberalism, Marxism, socialism, conservatism, etc.--i.e. purely a system of ideas, or an ideology. This way of looking at nationalism could not begin to explain its enormous emotional power, and its ability to make people willing to die for its sake.

This framework of the book helps to explain some typical forms of comparison that I employed, which were radically different from those used within the East-West format of 'The Idea of Power in Javanese Culture'. Where before I had been interested mainly in difference, this time I concentrated on similarity. The long chapter 'Creole Pioneers', on the Americas, is a good example. Most of the existing work on nationalism in the US either simply insisted on its exceptionality or linked it to British traditions. So I decided to compare the early US with the welter of new nationalisms in Spanish America and put it at the end of the chapter rather than at the start. I enjoyed anticipating the annoyance that would be caused by calling Franklin and Jefferson 'Creoles', as if they were simply an extension of patterns everywhere visible south of the US border, and commenting that Simón Bolívar was a more impressive figure than George Washington. In the same manner, I deliberately brought together Tsarist Russia with British India, Hungary with Siam and Japan, Indonesia with Switzerland, and Vietnam with French West Africa. (Many years later I enjoyed classifying Taiwanese nationalism as a late form of Creole nationalism.) These comparisons were intended to surprise and shock, but also to 'globalize' the study of the history of nationalism. Although I still like them, they are not much like the kind of comparisons done in mainstream'comparative government', which are usually based on statistics and surveys.

It was not until much later, in fact after I finally retired, that I began to recognize the fundamental drawback of this type of comparison: that using the nation and nation-states as the basic units of analysis fatally ignored the obvious fact that in reality these units were tied together and crosscut by 'global' political-intellectual currents such as liberalism, fascism, communism and socialism, as well as vast religious networks and economic and technological forces. I had also to take seriously the reality that very few people have ever been 'solely' nationalist. No matter how strong their nationalism, they may also be gripped by Hollywood movies, neoliberalism, a taste for manga, human rights, impending ecological disaster, fashion, science, anarchism, post-coloniality, 'democracy', indigenous peoples' movements, chat-rooms, astrology, supranational languages like Spanish and Arabic, etc. My realization of this serious flaw helps to explain why my Under Three Flags: Anarchism and the Anti-colonial Imagination (2005) focused not only on global anarchism towards the end of the nineteenth century, but also on global forms of communication, especially the telegraph and the steamship.

Because my framework had now changed, so did the style of the comparisons. Although 'The Idea of Power in Javanese Culture' and Imagined Communities were very different works, they had in common a strong longitudinal thrust. In the former, the reader moves across three centuries of Javanese history, while in the latter she is taken from the invention of print-capitalism in the fifteenth century to the anti-colonial movements of the mid-twentieth. In Under Three Flags the dominant impulse is latitudinal. The basic time-frame is marked, not by centuries, but by decades, just four of them between 1861 and 1901. What interested me most

was how political and literary developments such as anarchism and avantgarde writing were visibly linked, in what Walter Benjamin called 'homogeneous, empty time', in Brazil, Cuba, the UK, Belgium, Italy, France, Spain, Germany, Russia, South Africa, Japan, China, Oceania and the Philippines.

This kind of study required a new kind of narrative structure, more like that of a novel serialized in a newspaper than the ordinary type of scholarly historical work. The reader is invited to leap back and forth between Naples, Tokyo, Manila, Barcelona, Paris, Rio de Janeiro, Brussels, St Petersburg, Tampa and London. The emphasis is on contemporary learning, communications and coordination in connection with ideologies and political activism, thanks to the speed of telegraphic communication across state and national boundaries. Some Frenchmen were learning from some Americans and Belgians, some Chinese from some Filipinos and Japanese, some Italians from some Spaniards and Russians, some Filipinos from some Germans and Cubans. And so on.

While the general stress was on simultaneity and similarity, nonetheless the core of the book is an analysis of the contrast between global anarchism and local nationalisms. The nicest emblem for this contrast emerges from an investigation of the big wave of assassinations during the period, stretching from Buffalo, New York, to Harbin, Manchuria. Nationalist assassins always tried to kill 'their own' hated state leaders, while anarchist assassins very often targeted not only their local oppressors but notorious political leaders in

other countries.

It is important to recognize that comparison is not a method or even an academic technique; rather, it is a discursive strategy. There are a few important points to bear in mind when one wants to make a comparison. First of all, one has to decide, in any given work, whether one is mainly after similarities or differences. It is very difficult, for example, to say, let alone prove, that Japan and China or Korea are basically similar or basically different. Either is possible depending on one's angle of vision, one's framework, and the conclusions towards which one intends to move. (In the jingoist years on the eve of the First World War, when Germans and Frenchmen were encouraged to hate each other, the great Austro-Marxist theoretician Otto Bauer enjoyed baiting both sides by saying that contemporary Parisians and Berliners had far more in common than either had with their respective medieval ancestors.) In the present chapter I have thus tried, as perhaps offering a useful example, to show how the comparative works I wrote between the early 1970s and the 2000s reflected, in their real difference, changing perspectives, framings and (political) intentions.

A second point is that, within limits of plausible argument, the most instructive comparisons (whether of difference or similarity) are those that surprise. No Japanese will be surprised by a comparison with China, since it has been made for centuries, the path is well trodden, and people usually have their minds made up beforehand. But a comparison of Japan with Austria or Mexico might catch the reader off her guard.

A third reflection is that longitudinal comparisons of the same country over a long stretch of time are at least as important as cross-national comparisons. One reason for this has to do with the power of a certain kind of textbook-style national history that does not disdain myths and has a vested interest in continuity and perpetuating an ancient 'national identity'. Scots who want to believe and insist that they have long been oppressed by the English do not like to be reminded that London was ruled by a Scottish dynasty through most of the seventeenth century; likewise many Japanese do not take kindly to the suggestion that their country's earliest 'emperors' may have been partly Korean in origin. Hence scholars can profit immensely by reading widely in ancient history.

A fourth point is that it is good to think about one's own circumstances, class position, gender, level and type of education, age, mother language, etc., when doing comparisons. But these things can change. When you start to live in a country whose language you understand barely or not at all, you are obviously not in a good position to think comparatively, because you have little access to the local culture. You feel linguistically deprived, lonely and even isolated, and you hunt around for some fellow nationals to stick with. You cannot avoid making comparisons, but these are likely to be superficial and naive. But then, if you are lucky, you cross the language wall, and find yourself in another world. You are like an explorer, and try to notice and think about everything in a way you would never do at home, where so much is taken for granted. You can no longer take your class position, your education, even your gender, for granted. What you will start to notice, if your ears and eyes are open, are the things you can't see or hear. That is, you will begin to notice what is not there as well as what is there, just as you will become aware of what is unwritten as well as what is written. And this works both for the country you are living in and the one from which you came.

Often it starts with words. Indonesian, for example, has a special word, gurih, for the taste of rice ('deliciously pungent' according to one dictionary). If you come from England, you are then startled to realize that the taste of rice can't be described with a designated English word. On the other hand, Indonesian has no word like the English 'sepia' for the beautiful colour of old photographs. The same is true of concepts. Javanese has a word, longan, for the empty space under a chair or bed, which English does not.

Such a period of struggling with a new language is especially good for training oneself to be seriously comparative, because there is not yet any automatic lovely translation of foreign words into the language in your head. You gradually get to know enough to notice more, and yet you are still an outsider. If you then stay on long enough, things get taken for granted again, as they were back home, and you tend to be much less curious and observant than before. You start to say to yourself, for example, 'I know Indonesia inside out.' The point being that good comparisons often come from the experience of strangeness and absences.