City University of New York
Brooklyn College School of Education
Brooklyn, New York 11210 USA
In The World on Paper David Olson attempts to re-open the debate over how literacy influences the ways we think. He constructs a new theoretical object "literate culture" by writing a history of literacy practices in the context of wider social and cultural changes. He interprets this history psychologically, as changes in the forms of human mental processes caused by changes in literacy practices. His evidence is drawn mostly from secondary sources (historical analyses), and his own readings of a few primary ones; the argument dips into history at long separated moments, from the ancient Sumerians and Hellenes to the early moderns of 17th century Europe. He supplements this history with some developmental studies and a few ethnographic ones. Throughout, Olson always seems to seek the maximum generalization, even when more limited and modest hypotheses might be more convincingly sustained. In doing so, he psychologizes cultural history rather than attempting a genuinely cultural and historical psychology. Nevertheless, he gathers together in one coherent presentation a very large and complex subject; one can profit as much by disagreeing with what The World on Paper says as by assenting to it.
The Great Debate
The book's first agenda is to re-open the debate of the 1960s and 1970s over the role of literacy in cognition (e.g. Goody & Watt 1963, Havelock 1963, 1976; Ong 1976, 1982; Olson 1977 vs. Cole et al. 1971, Scribner & Cole 1981, Greenfield 1983). In her review of Scribner & Cole (1981), Greenfield concluded that they had settled the debate and "rid us once and for all of the arrogant and ethnocentric notion that a single technology [i.e. literacy] suffices to create in its users a distinct, let alone a superior, set of cognitive processes." There was passion in this debate!
The literatists had argued that by committing language to stable written forms some cultures had acquired the ability to: (1) re-examine the same text again and again, thus defining the notion of text-as-such independent of context; (2) compare texts and parts of texts, thus segmenting words and making grammars possible, and, most importantly in their eyes, (3) making possible in addition the conception of a context-free logic. Havelock argued this for the Greeks; the others developed the thesis for us moderns in comparison to contemporary non-literate cultures. Olson was particularly interested in the developmental progression from pre-literate to literate forms of reasoning in children.
The culturalists, on the other hand, pointed out that it is extremely difficult to separate the effects of literacy as such from the culturally specific forms of thinking produced by education, and particularly schooling, in modern societies. They examined a particular society, the Vai of Liberia, in which the existence of an indigenous script and literacy tradition not associated with schooling in European literate culture (it was used mainly for personal letters) made it possible to separate literacy as such from the effects of modern education. They found that the kinds of changes the literatists predicted (e.g. greater facility with decontextualized, even contrafactual syllogistic reasoning) were associated not with literacy as such (i.e. not with both the indigenous Vai and the imported European literacies), but only with the effects of eurocultural schooling.
Data have never yet killed a theory that didn't want to die, nor should they. Data and argumentation exert pressure on theories, which change to accommodate them. Theories die only of disuse, when other more useful models supplant them. In the face of the work summarized in Scribner & Cole, the literatists have shifted their position (see also Goody 1986, 1987). It is no longer literacy-as-such which is said to influence our thinking, but the writing, and especially the reading practices of a literate culture. The effect is due to schooling, but, they would now claim, what schooling teaches is itself the product of a history of progressively more sophisticated literacy practices, which are in turn the indispensable tools of higher rationality. These tools have evolved as they have, it is claimed, because of the nature of writing as a medium distinct from spoken language.
The World on Paper now cites the findings of Scribner & Cole (1981) several times as evidence for this neo-literatist position, and claims that the hypothesis which was falsified by them was only Scribner & Cole's own "naive" (p.41) interpretation of the literatist theory. Their work still stands, however, for most of us, as a pioneering demonstration of how the kinds of reasoning which eurocentrism projects as universally valid are in fact intimately interdependent with other social and cultural practices in a community. If this is true for non-European cultures, we can only expect that it is true for us as well. Our logic may have been made to be independent of the context of situation in which we use it, but it is not independent of the wider culture which contextualizes its use and meaning. The ways in which we situate our reasoning in its contexts are themselves culture-specific. The belief in universal context-independent principles of reasoning is itself specific to particular cultures, including our own.
For Olson, however, our logical principles are universal truths which European culture has "recognized", and he wants to construct for us the history of how we managed this great achievement. The neo-literatist position changes the terrain of the debate from the competence of individual subjects to the history of cultures. Literacy, in the sense of being able to read and write, is no longer the issue; the issue is whether or not, and how, literacy throughout history has shaped the cultural forms of our reasoning. I do not think many would deny that specialized reading and writing practices in particular registers and genres are associated with specific modes of meaning-making, or that these associations are the product of history and differ from culture to culture. But the neo-literatists want to say much more than this. Olson claims that literacy-derived modes of reasoning are general across all uses of representations in modern European culture (chaps. 8-11); that they have their origins in the nature of writing as a medium (chaps. 4-5), and as such, across cultures, they represent superior achievements wherever they occur (chaps. 6-7, 12).
Although I can see much of interest and value in many of the specific points which Olson makes along the way as he constructs his arguments for these grand theses, ultimately I do not find the theses themselves tenable in the form in which he presents them. Much of my own work in the last 15 years (e.g. Lemke 1983, 1985, 1988, 1989, 1990, 1992, 1995a, 1995b) and that of others in the traditions of text-based, discourse linguistics and social semiotics on which I draw (e.g. Halliday 1978, 1985, 1989; Martin 1992; Thibault 1991; Hodge & Kress 1988; O'Toole 1994; Kress & van Leeuwen 1990) has been concerned with how people make meaning with written text, spoken language, and most recently visual representations. This work, too, is interested in the social, cultural, and even historical (e.g. Halliday & Martin 1993) connections of our meaning-making practices. It comes to very different conclusions than Olson does.
Writing, Speech, and Meaning
Central to Olson's argument is the thesis that writing does not directly represent the illocutionary force of utterances (roughly, how they are to be taken; see below), and that it is the attempt to write and read so as to make authorial intentions recoverable which drives the history of literate culture. Writing does not represent illocutionary force (which for Olson is the same as authorial intent; a dubious identification, see below) mainly because it does not transcribe intonation.
Olson makes a very unfortunate oversimplification at the beginning of this argument when he says that "whereas spoken utterances tend to specify both what is said and how it is to be taken, written ones tend to specify only the former." (p.91). He grounds his argument here in functional linguistic theories such as those of Chafe (1985) and particularly Halliday (see references above). Their models are ones in which syntax is motivated by, or at least accounted for, in terms of semantic functions: wordings serve to construct meanings.
In Halliday's model, as it has been developed in the last two decades, there are three general functions which the resources of language as a semiotic system help to accomplish: (1) an ideational, thematic, or presentational function (saying the what about the world: what processes, events, and relations among what actors and objects, in what circumstances, etc.); (2) an interpersonal, interactional, attitudinal, or social-orientational function that lets us construct relations of speakers to audiences, and of speaking voices to their own ideational content and to other other possible social viewpoints (cf. Bakhtin 1935 on "heteroglossia"); and (3) a textual, structural-textural, cohesion-building, information-organizing function that tells us "what goes with what" to make larger unities and patterns.
Although there is a clear tendency for speakers to associate certain linguistic resources exclusively with certain functions (transitivity with the ideational, mood and modality with the interpersonal, reference-chains with the textual) during the earlier stages of language development, in mature discourse, both spoken-conversational and written-analytical, all the resources work together in complex register- and genre-specific patterns to fulfill all the functions (cf. Halliday 1985 on grammatical metaphor; Martin 1991, 1992; Hasan 1995; Lemke 1988, 1989, 1990, 1992, 1995a, 1995b).
Olson isolates a single resource, intonation (itself usually inseparable from prosody), which contributes in mature discourse to all these functions, but which has special cultural salience for English-speakers as a sign of "illocutionary force." Austin's (1962) term refers to the speech-act type of an utterance (e.g. whether it functions as statement or question, command or request, threat or promise, jokingly or ironically, etc.). It represents one aspect of the interpersonal or social-orientational function. Noting that intonation is not directly coded into written texts the way phonemic (alphabetic) information is, except for some conventions of punctuation, Olson argues that therefore what is fundamentally distinctive about written text is that, compared to spoken language, we have to do far more interpretive work to construct the social-orientational dimension of its meaning than we do for presentational-thematic or structural-organizational meaning.
This analysis is inadequate for a number of reasons. First, the social-orientational meaning of a text or utterance is coded in multiple ways, many of which are represented in written text, including word-choice and relations to preceding and following text in terms of conventional macro- and mini-genres and rhetorical formations (e.g. problem-solution, question-answer, challenge-response, etc.). The mood of verbs and their modality and modalizations contribute to these meanings, and so do a whole host of other grammatical and lexical options which are as visible on the surface of the text as are words themselves (cf. Halliday 1985; Martin 1992; Lemke 1988, 1989, 1990, 1992; Poynton 1989). The mere lack of visual coding of intonation patterns only produces about as much ambiguity in this aspect of meaning as one finds in the other two aspects (thematic and cohesive), particularly when we examine utterances in their discourse contexts.
Let's take an example from Olson (p.185), which he uses to illustrate his basic claim that written language deploys richer resources for indicating social-orientational meaning in order to compensate for the absence of intonation:
We read "He insisted that the book was his" without mentally "hearing" the insistent tone. Lexis comes to bear more and more of the tone.
By "lexis" here, Olson is pointing to the role of the word insisted. It certainly seems true that we do not ordinarily read the subordinate (or projected, Halliday 1985) clause here with any specially insistent intonation, but what is missing in Olson's analysis is an awareness that here insisted is the main or modal verb, the one which, in a proposition, is arguable. What is at issue in terms of social-orientational meaning here is therefore whether he insisted or not. Compare the typical "He insisted the book was his, didn't he?" with the grammatically desperate "He insisted the book was his, wasn't it?" to see this. If we elaborate on the modality of this main verb, as in "I regret to say that he may just possibly have insisted that the book was his," I think we will hear, not the insistence, but the regret and uncertainty, and the textual basis for this is a locution typical of spoken language and not at all special to written conventions.
This focus changes if we shift from subordination (hypotactic projection) to co-ordination (paratactic projection): "He insisted, the book is mine!" where the exclamation point now seems called for, and the intonation is certainly "heard" in the written text. In this case both verbs are focal. As Bakhtin-Voloshinov shows (1929/1986), such matters are very complex in languages, like Russian, where there are many sorts of semi-indirect discourse options in the grammar. Finally, note that in spoken language intonational information is normally redundant with many other cues that are reproduced in our written script: "I'm telling you for the very last time that the god-damned book is mine!" Try writing this sentence without the exclamation point to see whether it is just the punctuation that conveys the tone in writing. Compare this form of written expression, more typical in narrative fiction or dramatic dialogue than in academic prose with what Olson might consider the more highly evolved written form: "He insisted in the strongest possible terms that the book was his." We are not dealing here with general differences between spoken and written language, but with historically and culturally specific conventions of style in particular registers and genres.
Even beyond all this, Olson's view that "illocutionary force" is a matter of speaker intention is a very particular and narrow theory of social-orientational meaning. One can readily construct a purely formal-relational, functional model of this dimension of meaning just as one can for the thematic and organizational dimensions (see references to Halliday, Hasan, Martin, etc. above). It is simply our own folk-theory that associates this aspect of meaning more specifically with authorial or speakers' mental states than it does for the other two aspects. We do so mainly on the basis of our beliefs that ideational meaning can be set in correspondence with the world, and organizational with the text itself, leaving the orientational in correspondence only with the mental state of the speaker. It is very clear, however, in detailed analyses such as those just cited that all three dimensions of meaning work together to determine all three sorts of correspondences (if we accept a correspondence model at all). In semiotic and constructivist models of meaning, of course, world-states, texts, and mental states are themselves indeterminate apart from their meanings for us, which are made in part through language itself.
So it is not tenable either that intonation in itself makes so great a difference (as in the example above), or that "illocutionary force" is more indeterminate than textual meaning in general.
The next step in Olson's argument centers textual interpretation around the problem of recovery of "authorial intention." If the notion of authorial intention is useful at all, however, it cannot be identified so narrowly with "illocutionary force" or even with the whole of social-orientational meaning; authors have to have intentions with regard to the thematic and organizational dimensions of their meaning as well, and in practice the analytical separation of these dimensions from one another is not the sort of thing anyone except discourse analysts would normally do. Literary criticism has long since gone beyond models of the recovery of authorial intention as the central problem in the interpretation of texts, and so have linguistic and semiotic theories of text meaning. Only commonsense philosophy and cognitive psychology seem to lag behind in this.
Textual interpretation today is more often seen as the problem of constructing a useable meaning (not "the correct" meaning) by some practices which are arguably legitimate in the community. If you can get the usefulness of your meaning recognized, you then have a warrant even for non-canonical reading practices. Were the author to show up and state his or her "original" intention, we would add it to the list of available meanings of the text, not cross off all the other interesting interpretations. We would probably also interrogate the author to discover if perhaps s/he may have had more than one meaning in mind at the time of composition, and even ask again year by year if, living now in our time and culture, s/he did not find still other useful meanings to add to our list. As an author myself, I find it often takes but a few days, or having read a new book meanwhile, to go back and make a different reading of my own earlier notes or draft.
Authorial intentions today are regarded neither as monolithic (we are multiplex Selves who make multi-layered meanings) nor as stable (who can reconstruct with certainty the meaning we meant a month ago?), certainly they are not privileged -- except in certain genres for purely practical purposes, such as when transacting with a speaker or writer directly.
Perhaps this special case shows us the origin of the intentional fallacy in interpretation, as we can now call it. If I ask a waiter for some dish on the menu and he brings the wrong one, my interpretation of the meaning of my request should be privileged, though it may still have to be affirmed by the testimony of witnesses and an appeal to power. This privilege is part of the (idealized) norms of this interactional genre in my community. This special case, however, gives no warrant for extending the privileging of authorial intention to all genres and all uses of texts and language.
Interpretation, making meaning with text, is a matter of contextualization: we must always make sense of this text in the context of some other texts. We connect these texts through particular kinds of relations we construe among their patterns of thematic, attitudinal, and organizational meanings. We understand the ideas, attitudes, and unities of this text because we are familiar with similar or contrasting ideas, attitudes, and patternings in other texts. This goes on equally in spoken and written discourse, varying in its details more by register and genre than according to medium as such. Interpretation in this model is selective contextualization, and it is the reading practices of a community, far more than any feature of the text-on-paper, which specify acceptable meanings.
Olson never considers why no culture seems to have evolved a script that represents intonation or speech-act force directly; at least none seem to have gone beyond devices like the question mark. One reason might be that it is not necessary; that the plethora of redundant codings of social-orientational meaning by other means is good enough, in the discourse context, to do the job. Is it not strange that no script has a symbol for jokingness or ironic intention? The emoticons used by some communities on the Internet are such a notation, and in brief communications in quasi-conversational registers and genres they seem called for sometimes. But literary writers seem to have managed quite well without them; in context we can generally tell, when we are meant to be able to, whether a written line is to be taken as said with a smile or not.
There are of course today several technical notations for prosody and intonation in the transcription of speech. They are needed in the transcription of speech because its registers and genres did not evolve to rely more exclusively on the other interpersonal resources, since intonation is also available. Olson is quite right at least that in written genres we must look elsewhere for this information, and that many written genres have evolved specific ways to supply it. There is nothing special about this, however, in regard to social-orientational meaning: intonation plays a role in ideational-thematic meaning and in organizational meaning as well. Which aspects of the meaning of a text, spoken or written, require more text-specific work, and which can be handled by generic reading practices is register- and genre- and situation- specific; it is not specially a matter of spoken vs. written language as such.
Non-intonational resources for the indication of social-orientational meaning (attitude to addressee, attitude to content, attitude to other social opinions) are not more elaborately deployed in many written registers and genres than they are in many spoken ones. Scientific and technical texts are pretty bare in this respect; warrantability of claims and desirability of the investigation itself, the default options of the register, predominate and little special interpretation is needed. Literary texts and humanistic argumentation fit Olson's model better. But spoken language, even with intonation available and in use, still co-deploys vast additional resources to mark nuance in this domain and does so, it seems to me, very frequently, redundantly, and with great subtlety and elaboration. Face-to-face human relations, which are negotiated in these registers and genres, seem continually to require this kind of work, as much or moreso than do many written texts. Intonation alone would hardly be enough to do this important job in spoken language.
The Cult of Representation
The second strong claim in The World on Paper has to do not with the supposed peculiarities of the written medium, but with a purported direct extension of interpretive practices for texts to interpretive practices for Nature and for visual media such as scientific drawings and maps. If the claim were only that in the early modern period (roughly the 16-17th centuries) in European history, there were interesting parallels and connections among text-interpreting practices, early modern scientific interpretations of the "Book of Nature", and forms of genre painting and botanical illustration, we would all consider this plausible and take an interest in the specific details. But Olson argues that the changing literacy practices were originary and causal, rather than merely part of a complex network of reciprocal interrelationships.
Olson situates the origins of a concern with representations as distinct from the things represented in the historical development of hermeneutic practices for interpretation of the Christian sacred texts (see below). The construction of a notion of literal meaning which he finds there, he then sees as the source of a literalness about Nature in (mainly British) empiricism and the rise of early modern science. A book cited by Olson, Shapin and Schaffer's (1985) work on Boyle and Hobbes, tries to show these connections in some detail, though with more concern for the polarization of political vs. scientific issues within the context of a common religious worldview. Their work is the basis of Bruno Latour's brilliant essay We Have Never Been Modern (1993), which tries to show how the very separation of Nature from Culture in this period produced the modern worldview and what the original political and hermeneutic contexts of this disjunctive move may have been.
The account as we have it in Olson, based on several secondary sources and some selective quotation from Francis Bacon, seems to me a rather substantial oversimplification of the complexity of these issues. In particular, Olson seems to agree with some writers who project back onto the Early Moderns a more fully developed epistemology of representation than perhaps we should give them credit for. The views Olson attributes to them seem to me to be more those which we can only formulate from the time of C.S. Peirce or de Saussure. Foucault's (1966) account of the medieval to early modern transition gives both a more credible, and generous, reading of the former's reasoning through systems of correspondences and a less anachronistic view of the early moderns.
Olson's discussion of the history of maps and navigation is fascinating if you haven't heard these things before. He takes this as an example of the great progress that became possible once representations (maps) and the practices of creating and using them (cartography, navigation) became "freed" from dependence on immediate situational contexts (sightings and voyages reported, location of the map in the world). He shifts here, however, from a historical to an ethnographic account of non-literate navigation in the Caroline Islands of the South Pacific, implicitly identifying features of totally unrelated cultures (our past, their present), and explicitly marking our practices as superior, despite the fact that Carolinian navigation gets them where they are going across open ocean just as our does for us. It is the mere fact of ours being based on decontextualized representations that makes it "an advance."
I think one could well make the counter-argument that, compared to European methods, Caroline Islander navigation is more cognitively complex, more situated in the environment, at least equally theoretically decentered from ordinary experience (if that is an advantage), and certainly requires more active construction by the navigator and less routinization of procedures whose basis is not understood by the practitioner.
Bruno Latour (1987, 1993) has also discussed and analyzed these issues cogently and he makes the case that both procedures are equally situated in context, the difference being only that the European mode has a cover-theory that makes it seem more universal when it is still only the everywhere-local operation of a greatly more extended network of situated activities. We can certainly agree with Olson that the moderns sought to co-ordinate observations made in distant places by creating a scheme that highlights only those features of the observations that are readily portable and combinable. European procedures are mediated by maps and navigational aids and by the canonical practices that tie them together. Carolinian procedures are also mediated by practices tied together in complex networks, but perhaps tied together in a less analytically separable way; theirs is a more holistic situated practice that does not try to represent itself as anything more.
Olson's account of the mathematical-geometrical representation of motion, and so of time as analogous to space, in Galileo is a fairly standard one, but it does not see many profound issues having to do with how verbal and mathematical-visual forms are combined, or how the latter evolve away from the semantics of the former as they attempt to cope with phenomena of continuous change and ratio-of-ratios relationships. There is certainly a new mode of representation developing here, but it is not so much new visual forms (Newton's and Galileo's diagrams look pretty much identical to Euclid's) as it is a new way of combining the semiotics of natural language with that of geometry and mathematical notation.
There is something of the same problem with Olson's analysis of the history of literate culture (see below); literacy always existed in the context of oral traditions and spoken language practices, and meanings must have been made by the combination of the two, but Olson's history, by contrasting speech and writing, ends up misconceiving literate practices as essentially autonomous.
To link this all back to his main argument about textual interpretation, Olson adds in fictional representation as another parallel case. Modern fiction, he notes, presumes a distinction drawn in a particularly modern way between fiction and non-fiction. A medieval allegory is not told as if it were simply and literally true; Defoe's account of Crusoe is, and it owes much to the scientific empirical prose accounts of the early modern period.
Modern fiction for Olson operates with the discourse of pure literal representation. But again this is true only of certain genres, and they are extreme cases in which we do not know the illocutionary force, or the genre-identification fools us. Even allegory is surely not misplaced in the reading of many of these texts. Modern fiction strives for (or plays against) a verisimilitude that has little to do with literal journalistic Truth. I do not doubt that there are genres of fiction that rely on the earlier invention of the genre of the empirical scientific account, but I suspect that one could trace this interplay between factual genres and fictional ones back a lot further than the early modern period. Histories, travelers' accounts, diaries and biographies all had their fictional progeny in earlier ages. If medieval literature was predominantly allegorical rather than pseudo-factual (was it?), this was more likely the consequence of a cultural worldview in which allegorical meanings were naturally everywhere, inherently reliable, and extremely important.
Olson's concluding attempt (p.231) to fuse the fictional case and the scientific one is no more persuasive. The common ground, he claims, is that both scientific prose and literalist fiction took the ordinary reader-observer as their audience, not presupposing any esoteric intertextual knowledge for the deep interpretation of the hidden meanings of their texts. This may have indeed been the view of some writers at the time, but all fiction (all text) requires intertextual contextualization to be meaningful, and scientific accounts require rather specialized knowledge. At best, these writers (Boyle, Hooke, Hobbes, Defoe, Swift) wrote to a small audience of gentlemen with whom they could assume the necessary interpretive knowledge was already shared. They did not write to be read by workmen, washerwomen, children, or those they would have termed "primitives".
If the similarity seems partly well-founded for the empiricist-observational accounts of early modern science (Hooke or Boyle), it certainly does not for the theoretical-logicist prose of Galileo or Newton. A real Simplicio would no more have followed Galileo than a real Meno would have understood the genre of Socrates' demonstrations. Logical relations of pure abstractions are not propounded in the same literalist frame-of-reference as descriptions of herbs or of actual experiments. People who operate the literalist hermeneutic to make sense of the latter are quite lost as readers of the former. Once again we find a genre- and register- specific issue, not a universal paradigm of literate reading, not even a culturally and historically specific one.
The Pilgrims Make Progress
It rings a bit strange at the end of the 20th century to read a history which represents chronological sequence as objective Progress. Even stranger that it is a selective genealogy tracing the steps by which literate culture achieved its worldwide, transhistorical, transcultural pinnacle: the modern European hermeneutics of abstract representations, literal meaning, and authorial intentionality. The medieval period is again a time of superstition and chaos; the light is brought by Bible-reading Christians, particularly Protestants.
First, alphabetic literacy itself is assigned to the side of Light; it is associated with the Greek polis and Athenian democracy. For the first time anyone can learn to read; the monopoly of the scribes is broken forever. But the advance of Greek alphabetics over the earlier Semitic and middle-eastern syllabics seems trivial, a small adaptation to a less syllabically regular language. Athenian democracy was an oddity among the literate cities of Hellas, which collectively were a geographical backwater of Civilization, sandwiched historically between the literate and ruthless empires that came before and after. Advances in literacy have been most regularly associated historically with the rise of imperialistic and bureaucratic societies. Record-keeping made taxation more effective, armies better organized and supplied, legal discipline more standardized. Simpler systems of writing meant more scribes could be trained more quickly and more cheaply, extending the reach of empire, and no doubt lowering the status and wages of the necessary but potentially dangerous scribocracy. Even in Olson's own discussion of literacy in the middle ages we find an explicit argument by McKitterick (1990) that literacy then was a potent instrument of repressive political power, as we certainly know it has been in our own times.
I will not review Olson's long historical construction of the origins of modern interpretive practices. The key figures are theologians concerned with biblical hermeneutics. The argument begins from the thesis that while the medieval allegorical method of reading "could start interpretation; what it lacked was a clear means for stopping it or for privileging any one interpretation" (p.276). "Texts were seen as a boundless resource from which one could take an inexhaustible supply of meanings" (p144). The skeptic in me wants to wonder if maybe Olson here is projecting contemporary postmodernism back onto the premoderns. I suspect that there were very strict limits on the kinds of meanings that a text of a particular sort, especially a sacred one, could legitimately produce for the medieval reader. Much more limited than what would be possible today. Usefulness or interestingness in any of myriad subcultural contexts was not then the criterion of legitimacy. In this case something like consistency with christian spiritual tradition was.
Olson's genealogy traces developments in this small subcommunity (how many people could read the Bible even after Luther?) to a time when "the meaning of a text is austerely anchored in textual evidence". I suppose if it were not anchored it might be buffetted about by every passing social and cultural change, it might be available to legitimate many heresies potentially dangerous to economic and political interests, or just to theological or religious commitments. Then again we might see it not so much anchored as bound, imprisoned, disciplined, controlled. And by what or whom? Saying it was anchored by "textual evidence" seems rather objective, but the canons of what counts as such evidence, what the legitimate uses of intertextuality may be, how evidence may legitimately be marshalled, etc. must be set by certain reading conventions and institutional norms, and those in turn are set by people in disagreement with other people, people with differing interests as well as differing views. The very notion that there is only one correct interpretation creates the conditions of possibility for the control of interpretation and for domination through that control. It is not so clear to me that Protestant hermeneutics as Olson describes it was an advance. Could it not just as easily be seen as another contribution of literacy to social control?
To illustrate the issues at stake in interpretation, Olson provides a number of examples: To whom does an expression in the Bible refer? an Old Testament prophet or the predicted Christ? In the Declaration of Independence are all those created equal all homo sapiens or only all propertied European males? Strangely, remembering the basis for Olson's model of literate interpretation, I fail to see how any amount of intonational information not carried by the written text could help us resolve these potential ambiguities. The ambiguities themselves exist only insofar as we can contextualize the text differently, and they are mainly thematic-ideational ambiguities, which are resolvable by intertextual contextualization, if at all. They are certainly not issues of illocutionary force. If they are taken to be issues of the "literal meaning" of the text, then we are probably lucky that our culture has already added many alternative ways of construing textual meaning to our repertoire.
If they are taken to be issues of authorial intention (and ones that show rather clearly that this notion has to encompass all three aspects of meaning, not just the social-orientational), we have the problem that the examples given are both products of collective authorship; very likely even the communities which gave rise to these texts harbored some considerable internal diversity concerning what they "meant".
Olson is trying to write a history of literacy practices, and particularly of literate hermeneutics. He does convincingly show that for some registers and genres, in some subcommunities in Christian Europe, there were changes in these practices, particularly that medieval analogical and anagogical reading was reined in in some places by Protestant literalism. It is also plausible that in these same communities this literalism was part of a larger movement to separate object from subject, form from meaning, and text or observation from interpretation (cf. Latour 1993). Much of subsequent and especially recent intellectual history has been the story of our efforts to rejoin together what these early moderns, for their own purposes, put asunder.
Olson's argument, however, consistently contrasts literate modes with speech in ways that are linguistically doubtful as generalizations, however accurate they may be for particular written genres and registers. Grounding the contrast in universal, naturalized claims about the written medium makes the remaining analysis rather anti-historical in that it systematically ignores the specific historical interdependence of oral and literate modes of communication and meaning-construction. How can one argue that one of these, the literate, was the causal force or origin of later forms, when we do not have an analysis of how literate readings and oral traditions interacted with one another?
In most of Olson's arguments on the medieval period, there is really very little sense of joint practices combining the oral and the literate. Aries (1962) describes in great detail, and with some amazement, the career of Thomas Platter (b.1499), a Swiss, who for a decade studied across the length and breadth of Germanic Europe, acquiring the trivium and quadrivium, but only at the end actually learning how to read. At this point he found he already knew by heart most of the basic texts and he used his new skill only on previously unfamiliar ones. He opened his own school just a few years after learning to read; all his basic education in the literate cultural practices of his day had taken place without benefit of literacy as such. In this climate how can we credit analyses of literacy practices that ignore or are made in ignorance of the embedding oral practices? Even great texts such as Aquinas' Summa were written down, dictated, from prior oral composition. The great theological debates were oral debates, in the context of wider oral discussions and disputations, from which the written texts often came as after-effects. We cannot even conclude that the features of the medieval written text, its linguistic register features, were the product of literacy practices as such; they clearly descend from a thoroughly oral tradition that had a literacy adjunct.
All human cultures are oral cultures. Their written component has been tiny until very modern times, and it is perhaps with the bias of our modernist perspective that we tend to overestimate the influence of the written medium in other times and places.
Olson passes over the obvious need to compare his theses about European literate culture with the well-studied cases of China and India, two literate cultures of equal or greater antiquity. Only such a comparison could hope to establish scientifically the features of literate culture as such, and the effect of the written medium as such. The history of Chinese hermeneutics is hardly parallel to what Olson describes for medieval and modern Europe. In India, to a far greater extent than even in our Middle Ages, over a far longer period of time, oral traditions of scholarship and criticism were practiced with actual written texts playing a marginal or ceremonial role. Olson passes over these matters in a few words. His universalizing assumptions are not even checked historically against data on the literate cultures of Byzantium and Islam, which represent the direct continuations of the ancient classical traditions and the sources of Renaissance literacy. Critical historical consciousness today rebels against an ideological selectivity that can admit the Jews, belatedly, into our cultural genealogy (as Olson does) but refuses the Eastern christians (who belie the West European Renaissance claim to be the true successors of Greece and Rome, and refute the dear myth that classicism-plus-christianity leads to liberalism) and Islam, which was arguably the major civilizing influence on late medieval Europe.
Olson parallels this history of literate culture with a number of developmental accounts of how children today learn to get along in our modernist world. The two lines of argument converge at the end of the book in the claim that the greatest psychological advance made possible by literacy is the development of an awareness that humans have internal, intentional "mental states". This, along with the modern theory of representation, already discussed, represent, for Olson, the greatest influences from literacy on the history of cognition.
Olson takes a very realist position regarding mental states. We have them, in all times and places, because speech, through intonation, reflects speaker intention. Literacy only helps us become aware that we have them, and leads to the formation of an explicit theory of mind, by pushing us to see that when confronted by written text, where intonation and so the royal road to authorial intentionality are missing, we must reconstruct this mental reality explicitly. I have already expressed my skepticism regarding several of the links in this chain of reasoning. I have even more difficulty with the implied universality of a realist view of mind. Olson gives various accounts of the development in modern children of signs that they operate, even implicitly, with a theory of mind, i.e. assuming that they themselves, and others as well, "have" intentional mental states. These studies only seem to me to document that they are learning from the adults around them a cultural habit of construing experience in terms of the language of mental states. We learn to operate "as if" others had such states, because they operate this way, too.
Such phenomena, however, are hardly universal. Olson himself cites the example of the Homeric poems where there is no evidence for a language of attributed, or self-described, mental states. This despite the fact that elsewhere he accepts notions of deliberate deception, which certainly abound in Homer, as developmental evidence for an implicit theory of mind. The Homeric language is a very powerful tool for the narrative construction of character, which, if we follow Bruner (1991, Bruner & Weisser 1992) is a primary means for the construction of a sense of Self, but which, as in Homer, clearly need not include in every culture our European belief in mental states. Olson also notes, in a different context, the case of the Quechua, who also do not share our mentalistic model or its language to account for what we call "false beliefs" (they say that things often look like what they are not, not that we are wrong in our "beliefs" about them). There is also other counter-evidence in the anthropological literature (e.g. Rumsey 1990) to Olson's assumption (p.128) that all cultures make even the basic meta-linguistic distinction between saying and thinking required by his model.
What then remains credible in the major theses of The World on Paper? Not that literacy has led us to an appreciation of the reality of mind, but only that it has played a role in the evolution of our own culture's modern folk-theories of mental states and those of its reading practices which foreground authorial intention. Not that literacy has led to objective transcultural progress in our understandings and uses of semiotic representations, but only that in early modern European cultural history there were close connections among the ways in which these representations were assembled and used in some specific domains of science, art, literature, and textual hermeneutics. Not that our literate culture, or literate cultures generally, were primarily shaped by a written medium that neglected to transcribe illocutionary force, but only that within the context of primarily oral cultures, literate meaning-making practices in some particular registers and genres evolved specialized means of wording relations among the thematic, orientational, and organizational aspects of all discursive meaning.
Particular historically and culturally specific forms of literacy participate in a complex web of interdependencies among the material meaning-making practices of a community. In particular times and places, for particular written registers and genres, we can examine, where the evidence is available, how these practices interact with spoken language, with the "inner speech" we call thought, and with other semiotic practices such as depiction and meaningful physical action. With very great caution and intellectual humility we can attempt to write histories of these interactions, always remembering how much of the living webs of meaning-making in the communities of the past have left no record or trace down to our own time. The World on Paper does us the great service of ambitiously framing these tasks and beginning the interdisciplinary work of pulling together what we think we know that may be relevant to pursuing them. I have learned as much from The World on Paper where I dissented from its arguments as when I agreed with them. What matters is that we create a discourse about meaning-making as culture and through history. The World on Paper contributes to this grand project of our own culture in the present moment of its history.
Aries, Philippe. (1962). Centuries of Childhood. New York: Random House.
Austin, J.L. (1962). How to do things with words. London: Oxford University Press.
Bakhtin, M. [aka V.N. Voloshinov] (1929/1986). Marxism and the philosophy of language. Cambridge, MA: Harvard University Press.
Bruner, J.S. (1991). The invention of self: Autobiography and its forms. In D.R. Olson and N. Torrance, Eds., Literacy and Orality. Cambridge: Cambridge University Press.
Bruner, J.S. and Weisser, S. (1992). Autobiography and the construction of self. Cambridge, MA: Harvard University Press.
Chafe, W. (1985). Linguistic differences produced by differences between speaking and writing. In D. Olson, N. Torrance, and A. Hildyard, eds., Literacy, language, and learning, pp.105-123. Cambridge: Cambridge University Press.
Cole, M., Gay, J., Glick, J., & Sharp, D. (1971). The cultural context of learning and thinking. New York: Basic Books.
Foucault, M. (1966). The order of things. New York: Random House.
Goody, J. (1986). The logic of writing and the organization of society.
Goody, J. (1987). The interface between the oral and the written. Cambridge: Cambridge University Press.
Goody, J. & Watt, I. (1963). The consequences of literacy. In J. Goody, Ed., Literacy in traditional societies. Cambridge: Cambridge University Press.
Greenfield, P. (1983). Review of "The Psychology of Literacy by S. Scribner and M. Cole". Harvard Educational Review 53, 216-220.
Halliday, M.A.K. (1978). Language as social semiotic. London: Edward Arnold.
Halliday, M.A.K. (1985). An introduction to functional grammar. London: Edward Arnold.
Halliday, M.A.K. (1989). Spoken and written language. Oxford: Oxford University Press.
Halliday, M.A.K. & Martin, J.R. (1993). Writing science. Pittsburgh: University of Pittsburgh Press.
Havelock, E. (1963). Preface to Plato. Cambridge: Cambridge University Press.
Havelock, E. (1976). Origins of Western literacy. Toronto: OISE Press.
Hasan, R. (1995). The conception of context in text. In M. Gregory and P. Fries, Eds. Discourse in society: Functional perspectives Norwood, NJ: Ablex Publishing.
Hodge, R. & Kress, G. (1988). Social semiotics. Ithaca: Cornell University Press.
Kress, G. & van Leeuwen, T. (1990). Reading images. Geelong, Victoria: Deakin University Press.
Latour, B. (1987). Science in action. Cambridge, MA: Harvard University Press.
Latour, B. (1993). We have never been modern. Cambridge, MA: Harvard University Press.
Lemke, J. L. (1983). Thematic analysis: Systems, structures, and strategies. Semiotic Inquiry 3(2), 159-187.
Lemke, J. L. (1985). Ideology, intertextuality, and the notion of register. In J.D. Benson and W.S. Greaves, Eds., Systemic perspectives on discourse, pp.275-294. Norwood, NJ: Ablex Publishing.
Lemke, J. L. (1988). Discourses in conflict: Heteroglossia and text semantics. In J. D. Benson and W. S. Greaves, Eds. Systemic functional approaches to discourse, pp. 29-50. Norwood, NJ: Ablex Publishing.
Lemke, J. L. (1989). Semantics and Social Values. WORD 40(1-2), 37-50.
Lemke, J. L. (1990). Technical Discourse and Technocratic Ideology. In M. A. K. Halliday, John Gibbons, and Howard Nicholas, Eds., Learning, keeping, and using language, Vol. II, pp. 435-460. Amsterdam: John Benjamins.
Lemke, J. L.(1992). Interpersonal meaning in discourse: Value orientations. In M. Davies and L. Ravelli, Eds., Advances in systemic linguistics: Recent theory and practice, pp.82-104. London: Pinter.
Lemke, J. L. (1995a). Intertextuality and text semantics. In M. Gregory and P. Fries, Eds., Discourse in society: Functional perspectives, pp.85-114. Norwood, NJ: Ablex Publishing.
Lemke, J. L. (1995b). Textual politics: Discourse and social dynamics. London: Taylor & Francis / Falmer Press.
McKitterick, R. (1990). The uses of literacy in early medieval Europe. Cambridge: Cambridge University Press.
Martin, J.R. (1991). Intrinsic functionality: Implications for contextual theory. Social Semiotics 1(1): 99-162.
Martin, J.R. (1992). English text. Philadelphia: John Benjamins.
Olson, D. (1977). From utterance to text: The bias of language in speech and writing. Harvard Educational Review 47(3), 257-281.
Ong, W. (1976). The presence of the word. New Haven: Yale University Press
Ong, W. (1982). Orality and literacy. London: Methuen.
O'Toole, L. M. (1994). The language of displayed art. London: Leicester University Press.
Poynton, C. (1989. Language and gender: Making the difference. Geelong, Victoria: Deakin University Press.
Rumsey, A. (1990). Wording, meaning, and linguistic ideology. American Anthropologist 92(2), 346-361.
Scribner, S. & Cole, M. (1981). The psychology of literacy. Cambridge, MA: Harvard University Press.
Shapin, S. & Schaffer, S. (1985). Leviathan and the air pump. Princeton, NJ: Princeton University Press.
Thibault, P. (1991). Social semiotics as praxis. Minneapolis: University of Minnesota Press.
"Kembali ke daerah ini"
08 Oktober 2012,
Taman Ilmu & Budi