Conceptualizing the World: An Exploration across Disciplines

By Helge Jordheim

What is—and what was—“the world”? Though often treated as interchangeable with the ongoing and inexorable progress of globalization, concepts of “world,” “globe,” or “earth” instead suggest something limited and absolute. This innovative and interdisciplinary volume concerns itself with this central paradox: that the complex, heterogeneous, and purportedly transhistorical dynamics of globalization have given rise to the idea and reality of a finite—and thus vulnerable—world. Through studies of illuminating historical moments that range from antiquity to the era of Google Earth, each contribution helps to trace the emergence of the world in multitudinous representations, practices, and human experiences.

• Language: English
• Publisher: Berghahn Books
• Release date: Dec 17, 2018
• ISBN: 9781805394075

Book preview

Part I

NAMING THE WORLD

1

World

An Exploration of the Relationship between Conceptual History and Etymology

Ivo Spira

As a global keyword, world has translation equivalents in all major languages.¹ This fact may easily lead one to conclude that there is also a corresponding global concept world, one that has come to dominate global discourse. It would be hasty, however, to take the translatability of this common English word as conclusive evidence for the hegemony of a universal concept of European origin. Not only do translation equivalents have different ranges of meaning, but they may not even be conceptually identical in the relevant senses. Perhaps this incompatibility can be traced to differing etymologies. After all, the translation equivalents have different histories, sometimes radically so, each in its own language. Thus, Greek kósmos originally meant order (as in cosmic order) before acquiring the spatial and social sense of the world (we live in) in Hellenic Greek, which is very different from the history of English world, which originated as a compound of the Germanic bases of were (man, husband) and old (lifetime, age).² Are these differences in word history, then, evidence for the existence of two entirely different concepts, namely world and kósmos? If so, etymological differences may reflect differences in the way in which different people and cultures understand the world—that is to say, they could be anthropologically significant. And if it turns out that the differences are significant, can the various local or regional concepts still be meaningfully subsumed under a global world concept? Even if we postulate this universal concept of world (e.g., as the result of the repeated reinforcement of equations like Gk. kósmos = Eng. world), we are still left wondering whether it is legitimate to make the etymological differences responsible for any tensions we detect in the translation equations. In what follows, I show that while etymology is indeed a subtle and significant factor in the formation of key concepts, it is the historical moments of conceptual definition that determine their meaning and significance.

In this chapter I approach the problem outlined above on two levels. On one level, I look into the etymological origins of words meaning world in a number of languages, with the goal of seeing if and how they contribute to the perceived global concept world. On another level, I offer some reflections on more general questions: (a) How do key concepts arise historically? (b) Does a word’s etymology reflect or, alternatively, influence the way in which it conceptualizes something? (c) What is the value of etymological analysis in analyzing a key concept or writing its history? In order to put my brief exploration on a firmer footing, it is necessary to state clearly what I mean by etymology and concept, and, further, how the words to be compared have been selected.

When discussing etymology, it is important to distinguish the actual historical development of a word (historia ipsa) from various accounts or interpretations of this development (historia narrata). The former is never directly accessible and only approachable by means of the latter. We may further differentiate between scientific and non-scientific etymological interpretations. The former in principle adheres rigorously to historical-critical principles, arriving at a hypothesis of historical development through a systematic consideration of all the evidence in its historical context, relying on all that is known about language change. Of course, that does not in any way guarantee that the resulting interpretations will be historically accurate or even unbiased. They always remain hypotheses embedded in their particular research context. Non-scientific etymological interpretations occur naturally, as they help speakers make sense of the language they speak. Moreover, etymological interpretations of both kinds are useful rhetorically. In practice, the border between scientific and non-scientific etymology is often blurred, since scientific paradigms and available sources change over time, and also since etymological interpretations readily move between the two spheres. Hence it is often a good idea to consider a broad range of possible etymological interpretations, scientific as well as popular, whether the goal is to approach the actual historical development of word forms or simply to investigate the dynamics of form and interpretation.

By concept in the narrow sense, I mean a notion that is sufficiently clear to be definable; in an even narrower sense, each concept is determined by its one and only definition, which is the ideal for key concepts in technical texts, for example voltage. These are usually accessible exclusively through technical terms, and the only valid meaning is licensed by the relevant definitions.³ Human language, however, is characterized by polysemy, that is to say, words tend to have more than one sense, where senses are semantically and historically interrelated, yet discrete, semantic units (including concepts in the narrow sense).⁴ Since speakers habitually make use of the flexibility inherent in linguistic polysemy, one may want to treat a whole cluster of concepts (in the narrow sense) as one overarching super-concept. This happens when speakers exploit the polysemic potential to augment the meaning of a word, so that it can express this super-concept. This turns out to be a very useful way of thinking about social key concepts, because it captures the way in which social keywords actually function—namely as ambiguous symbols, through which the skilled politician may summon multiple definitions and associations according to the context and party utility at any given moment. This is all the more true in a multilingual context, where the concept comes to function as a sort of super-symbol that subsumes (and partly obscures) the differences between the particular concepts that figure in the various translation equations.⁵ It is this broader sense of concept as super-concept, or super-symbol, that I adopt in this chapter unless otherwise specified. Thus, each concept is treated as potentially multivalent semantically and pragmatically,⁶ much in the way Reinhart Koselleck characterizes his Begriffe, especially the Grundbegriffe, or key concepts.⁷

Turning now to the methodological question of how to identify the words that should be investigated, there are basically two avenues of approach. One approach, which may be called semasiological, is to use the modern super-symbol of the world as a starting point and select the translation equivalents that are available for this word in various languages, and then proceed to describe and compare their semantics. The other approach, which may be called onomasiological, is to start with a definition (a concept in the narrow sense), and then find words in different languages that have a sense corresponding to the definition. This procedure is then repeated for each definition that is relevant to an exploration of the super-concept world. I have resorted to a combination of both approaches in order to get different perspectives.⁸

The modern translation equivalents of English world include the following: Standard Chinese shìjiè 世界, tiānxià 天下, rénjiān 人間; Standard Arabic ‘ālam, dunyā; Norwegian Bokmål verden; Modern Greek kósmos; French monde; Russian mir, svet. This gives us a list of words that are similar enough semantically to function as translation equivalents. In order to investigate their in-language semantics in detail, we would have to go far beyond available dictionaries and look at a broad range of examples in context for each of the words, including, crucially, their occurrence in various kinds of idioms, technical terms, and collocations.⁹

This leads naturally to the other approach, where we can use the set of senses obtained for each of the words in the list to search for, and ask questions about, other words with one or more of those same (or similar) senses. In this way, one will quickly end up investigating whole word fields, so that we might want to consider universe, cosmos, globe, creation, nature, and Earth in addition to world, to take a few examples from English. Investigating the relevant word fields is in any case necessary to get a full picture of a word’s place within a language. Eng. world, for example, is synonymous with universe in the sense of the whole physical space and everything that is in it, but synonymous with Earth in the sense of the planet Earth, where we live.

We can further enlarge the sample of words to study by applying both of the approaches diachronically. In this way, we will for example find Latin mundus by looking at modern English translations of Latin texts from different periods, and Greek oikouménē and Chinese tiānxià 天下 by looking for words with the sense the known, inhabited world.

After these preliminary remarks, I can now turn to my main topic, which is to investigate the conceptual significance of etymology, or word history, taking English world as my point of departure. As a first step, can one identify any patterns in the historical processes of the conceptualization of world, moving from past to present, from the local to the global? If we find any patterns of conceptualization that are shared between languages, or if we find significant differences, we may venture to suggest that they are anthropologically important, as they reflect similarities and differences in the perception of the world and the construction of the world across cultures.

In a survey of words for world, obtained through the approaches just exemplified, it becomes readily apparent that words or morphemes with the senses age (as in Gk. ai n, Heb.‘ōlam), heaven (Ch. tiān in tiānxià 天下, lit. [all] under heaven), earth (Lat. terra), sea (Ch. h i in h inèi 海内, lit. within the sea), create (Eng. creation), or dwell (Gk. oikouménē) recur as constitutive elements across different languages. Combinations also occur, for example Chinese tiāndì 天地 (lit. heaven [and] earth, meaning universe) and Arabic al-samāwāt wa-l-’ar (the heavens and the earth) in the Qur’ān.¹⁰ We can also point to differences in the word histories for words that later become synonyms or translation equivalents. Chinese tiānxià 天下 refers to the inhabited world; its denotation thus overlaps with Greek oikouménē (lit. the inhabited [earth]), while the familiar Greek word kósmos (lit. order) only much later became a synonym of oikouménē via the notion of cosmic order.

It is tempting to conclude that these elements correspond to cognitive notions that are anthropologically significant in the human conceptualization of the world. Surely the semantic differences and similarities in word structure matter, since it is through these that we name, and hence understand, reality? The imagery and reasoning that underlie the form of words with which we choose to express crucial concepts thus seems to have anthropological relevance. In this light, C. D. Buck’s classic Dictionary of Selected Synonyms in the Principal Indo-European Languages seems a good starting point for exploring human perception and conceptualization of the world—and, as it happens, world is in fact the first synonym group of Buck’s dictionary.¹¹

There are, however, reasons to be critical of the admissibility of word histories as evidence for anthropological significance. To begin with, the claim as it stands is too vague. For example, anthropological significance could be attached to the way in which the name is derived from words for salient properties of the named entity (salient in the perception of the naming subject, that is); but such significance could equally well be attached to contingent circumstances of the act of naming (the name may have originated as a joke). As long as one is sufficiently clear about what one has in mind, this kind of vagueness probably does not represent a major obstacle.

We may begin by noting that although contingent circumstances certainly play an important role, there is much to be said for the case that speakers are constrained by the morphological composition and etymology of the words and phrases they use. Ultimately, this is because the requirement that communication be intelligible and socially acceptable will severely constrain speakers in exercising the power to change their speech at will. This power is in principle arbitrary and unlimited, since you can make up any number of new words or change the meanings you attach to old words whenever you want. But since changing one’s language all the time precludes effective communication, speakers in practice reproduce existing norms.¹² There is, in other words, no such thing as an effortless escape from the constraints imposed by the existing senses of linguistic forms. Since this means that the choice between alternative words is constrained by the meaning of their components, we are in principle entitled to derive significance from their morphological and etymological structure. This kind of analysis, however, is valid only as long as we take the historical circumstances of naming into account (a point I will return to below), and only as long as we do not insist on a scientific, historically accurate etymology as the source of significance, but focus instead on the etymology and structure of words as understood by the relevant speakers.

Because of changes in the form of individual words and their frequency of use, specific interpretations of word structure become more opaque to speakers with the passage of time. This naturally also gives rise to alternative interpretations of the same word (reanalysis). The increasing opaqueness of word forms relative to earlier analyses makes it more problematic to derive current conceptual structure from an unearthed (putatively original) etymological structure the further one gets away from the original naming event.¹³

This means that drawing conclusions about the current meaning of words and concepts from their historically underlying morphological composition is a very risky undertaking, which needs to be conducted with the utmost caution if it is to be attempted at all. The origin of word forms can only provide ancillary evidence for current word meanings or conceptual significance. The etymology of a word tells us which resources were marshaled to [[p32]]meet certain expressive needs, without telling us much about how or why. In particular, formal etymology alone cannot tell us how the different elements were combined in the historical naming event. It may in fact be far from a logical combination of the senses of its components.

Thus, we know that English world goes back to Old English woruld, which is closely related to Old High German weralt and Old Norse ver ld. This Germanic word is a compound of the elements found in Old Norse as verr (man, husband) and ld (age, lifetime). We can now try to sniff our way back along the trail to Indo-European * īros (man < the strong one) and the root *al- (grow; nourish). We then find other cognates of these roots in other Indo-European languages, such as Latin vir (man) and alere (nourish). We are now in a position to congratulate ourselves on having convincingly demonstrated the ancient pedigree of the English key concept world, concluding that it is essentially an ancient Indo-European temporal concept, from which we then attempt to draw wide-ranging anthropological conclusions.

However, it turns out that there is no simple correlation between the keyness of concepts and the history of the word forms that express them. Not only is there no Latin concept *viraltum, but whereas vir did become a key concept in Roman culture, the Germanic cognates of Old Norse verr (man) hardly have any other presence in Germanic languages except English werewolf, German Werwolf, and similar cognates. Coincidentally, there is a relevant Latin key concept, namely saeculum (generation), but far from being linked to woruld by virtue of being a cognate, saeculum, augmented by its subsequent biblical sense of worldly life, licensed the meaning of woruld as a key concept. In other words, saeculum served as a cultural and linguistic model for woruld. Here we see clearly that the formal etymological genealogy does not explain the formation of the concept, but that investigating possible models in authoritative texts goes a long way toward a plausible explanation. It is easy to give other examples of the divergence of formal etymological relatedness and conceptual significance. English Earth is clearly a prominent cultural keyword, while its cognate Greek era is marginal. The culturally central Greek word for Earth is gē (as in geography), of unknown provenance. In Latin we have for example tellus, which has the socially and culturally relatively insignificant Old Norse cognate þilja (plank, planking).

Instead of deriving linguistic forms and social institutions from their (reconstructed) Indo-European proto-forms, Émile Benveniste wisely focused on the emergence of the lexical systems of particular languages as correlates of the developments of social and cultural institutions. In his words, the elements inherited from the common language find themselves incorporated into independent structures which belong to particular languages; hence they [the elements] transform themselves and take on new values among the oppositions which are thus created and which they determine.¹⁴ This means that an etymological analysis, even if it is both scholarly sound and successful in approaching the actual historical development, does not automatically allow us to determine reliably the motivation behind its structure, or what the word meant or how it was actually used. To discover the meaning of a word, we need to consult a corpus of extant texts to see how it was used at a given point in time by certain speakers and, if we are lucky, how it was defined, interpreted, or contested. We are even luckier if we happen to discover pointers to the motivation behind a word’s structure in terms of an authentic and relevant account of how it was coined. In the end, it is only in the context of a wider historical and cultural analysis that we stand a chance of finding plausible explanations for the creation or adoption of words to express certain concepts.

In the case of Old Norse ver ld (lit. age of man) and its Germanic cognates, the source of their meaning is basically the Christian concept expressed in Medieval Latin as saeculum. But saeculum itself was modeled on Greek aiō n (lifetime, eternity), which in turn had been used to translate Hebrew ‘ōlam (age, era).¹⁵ Another word connected with ‘ōlam is Arabic ‘ālam (world), which may actually be a loanword from Hebrew or Aramaic. It is prominent in the Qur’ān, for example in the divine epithet rabb al-‘ālamīn (Lord of the Worlds), echoing the Hebrew ribbon ha-‘ōlamīm. In Qur’ānic discourse we also find (al- ayāt, al-dār) al-dunyā (lit. the lower/nearer [life, abode], that is, the world of our human, earthly existence), which contrasts with (al-dār) al-’ākhira (the last about, i.e., the hereafter), which is temporal in a way reminiscent of Hebrew ‘ōlam, as in ha-‘ōlam ha-zeh (this age, i.e., life on earth) and ha-‘ōlam ha-bā’ (the world to come).¹⁶ In Modern Arabic, dunyā in the sense of the world we live in is synonymous with ‘ālam, albeit with different connotations. But neither of them is necessarily used in a religious sense.

Let us take a different example. In Ancient China, the world where people live was called tiānxià, and this was largely taken to be co-extensive with what became the Chinese Empire. The most common modern equivalent in Chinese for English world is shìjiè 世界 (lit. generation border), which has its origin as a translation equivalent for Sanskrit loka in the translation of Buddhist texts from the second century CE onward. The modern meanings of the word arose when it was pressed into service as a translation equivalent of English world, French monde, and German Welt in the nineteenth century.

This contrast between a narrow etymology of word forms and a wider investigation of concept formation and conceptual influence within and across languages strongly suggests that the moment in which a word’s form and meaning is created, recast, or reproduced—typically by way of an authoritative external model—plays a decisive role in the emergence of keywords. The concept meaning thus ends up by transcending the etymology of the concept as a word. We may call this crucial moment of conceptual definition the formative moment of the concept within a given language, or perhaps more accurately, within a given speech community.

This, however, requires some qualification. First is the one just mentioned: the continued resonance of earlier senses. To the extent that people are aware of them, these older semantic layers may continue to color their perceptions of the new concept, regardless of whether this was the conceptual innovator’s intention or not. Second, the term in question does not become a vehicle for the concept for all speakers at the same time: there is a lag that depends on both social and technological factors.¹⁷ Only if the concept is successfully established will it come to dominate the word (term) that expresses it, and it may eventually trump some or all of its existing senses and etymologies. Third, on the level of scholarship it is important to note that an etymological investigation still has some value, as it shows us which semantic resources were marshaled to form the word(s) that ended up as a vehicle for the new concept. Since the choice of word forms is not arbitrary, etymological and morphological readings that were accessible to relevant speakers at the time of concept formation do have some explanatory potential.

As noted above, it is often the case that the model according to which a concept is formed is external to the language community. An interesting aspect of such impulses from the outside is that the burden of conforming to the existing lexicon, especially the established senses of words, seems to be considerably lessened when a word form is taken from another language (i.e., as a loanword). In one way, this follows naturally from the fact that the loanword is a new element in the recipient language, and as such does not have any interfering, older senses. In another way, the burden is lessened because the community of the source language has its own sources of meaning and authority, for instance in the form of knowledgeable persons, reference works, and sacred texts, which the recipient community can have recourse to. There may also be considerable gains in terms of status if the model language enjoys high prestige. Another reason why it is tempting to take in words from other languages is that one may be highly selective as to which senses are adopted. This is typical in the case of technical terms: the basic sense of nebula in Latin is mist, but in English it is exclusively used in the technical sense of cloud of gas and dust in outer space, and is not a general synonym of mist.

If an existing word is chosen, or a new one constructed from native building blocks, one will have to provide conceptual support in a different way (e.g., through translation, exegesis, or new reference works), and there may moreover be dissonant earlier senses. Older meanings may be activated by a suitable context, that is, a context that selects the older senses. The activation of different senses on different occasions contributes to a kind of semantic osmosis within the word. These older semantic layers may, however, actually be used to good effect in mediating the new (external, foreign) concept, and thus constitute an advantage. One of course also needs to keep in mind that even with an external conceptual or lexical model, the concept may in fact end up being remolded in its replicated form, for example through a redefinition or through other, more subtle forms of appropriation in the new speech