shytone  books  music  essays  home  exploratories  new this month

book reviews



John Lewis Gaddis: The Landscape of History:
how historians map the past
(Oxford University Press: 2002)


“We are born, each of us, with such self-centeredness that only the fact of being babies, and therefore cute, saves us. Growing up is largely a matter of growing out of that condition: we soak in impressions, and as we do so we dethrone ourselves - or at least most of us do - from our original position at the center of the universe.... [And,] if that’s what maturity means in human relationships - the arrival at identity by way of insignificance - then I would define historical consciousness as the projection of that maturity through time.... [However,] just as historical consciousness demands detachment from - or if you prefer, elevation above - the landscape that is the past, so it also requires a certain displacement: an ability to shift back and forth between humility and mastery....[Because] the best you can do, whether with a prince or a landscape or the past, is to represent reality: to smooth over the details, to look for larger patterns, to consider how you can use what you see for your own purposes. That very act of representation, though, makes you feel large, because you yourself are in charge of the representation: it’s you who must make complexity comprehensible, first to yourself, then to others. And the power that resides in representation can be very great indeed.... Historical consciousness therefore leaves you, as does maturity itself, with a simultaneous sense of your own significance and insignificance.... You’re suspended between sensibilities that are at odds with one another; but it’s precisely within that suspension that your own identity - whether as a person or as a historian - tends to reside. Self-doubt must always precede self-confidence. It should never, however, cease to accompany, challenge, and by these means discipline, self-confidence.”
(Gaddis, pp.5-8)

Those familiar with this site’s preferences will by now be well-aware of just how highly I rate the historical disciplines re humanistic understanding...for a myriad of reasons involving intellectual history, the inherent biases of our human nature, and a sensible preference for the time-tested facts of human complexity over the all-too-easily misleading (and necessarily simplified) toy worlds of theory. And, thankfully, there exists a strong tradition of historiography by historians - as opposed to philosophers - which is a marvellous resource for anyone seeking to deepen their understanding of this supremely human discipline via the ideas of some of its most distinguished practitioners.

However, despite the litany of famous names who have contributed to this genre, my favourite work by far is this brief (and very recent) work by the distinguished Cold War historian John Lewis Gaddis. And the reason for this may already be evident from the opening quotation above. Because, rather than simply focusing on the specialist activities of historians, Gaddis is intent upon using these to help explore our general nature at historical/narrative beings...and building bridges with the historical sciences in general. The result is a work both lucidly-written and profoundly thought, and essential reading for humanists (and humans) of all stripes...


“Tension between particularization and generalization - between literal and abstract representation - comes with the territory, I think, when you’re transmitting vicarious experience. A simple chronicle of details, however graphic, locks you into a particular time and place. You move beyond it by abstracting, but abstracting is an artificial exercise, involving an oversimplification of complex realities.... But [historians] can, by means of that particular form of abstraction we know as narrative, portray movement through time, something that an artist can only hint at. There’s always a balance to be struck, though, for the more time the narrative covers, the less detail it can provide.... This then, is yet another of the polarities involved in historical consciousness: the tension between the literal and the abstract."
(Gaddis, pp.14-15)

But perhaps the most startling evidence of Gaddis’ originality is his insistence on re-framing the very bases of history - time and space - in insistently humanistic terms...rather than necessarily deferring to physics. For, as he explains, for our purposes here, the fundamental “truth” of such matters is far less relevant than the truths of our experiences of same (albeit tempered by the particular insights historians have patiently accumulated and debated). And so, the reversible time of most physical theory is an irrelevance...for the metaphors we need are those which illuminate our condition:


“I prefer to think of the present as a singularity...through which the future has got to pass in order to become the past. The present achieves this transformation by locking into place relationships between continuities and contingencies: on the future side of the singularity, these are fluid, decoupled, and therefore indeterminate; however, as they pass through it they fuse and cannot be separated.... By continuities, I mean patterns that extend across time. These are not laws, like gravity or entropy; they are not even theories, like relativity or natural selection. They are simply phenomena that recur with sufficient regularity to make themselves apparent to us.... We might define the future, then, as the zone within which contingencies and continuities coexist independently of one another; the past as the place where their relationship is inextricably fixed; and the present as the singularity that brings the two together, so that...history is made.”
(Gaddis, pp.30-1)

“If there is such a division for space, I suspect it lies in the distinction between the actual and the cartographic. The making of maps must be as ancient and ubiquitous a practice as is our three-part conception of time. Both reduce the infinitely complex to a finite, manageable, frame of reference. Both involve the imposition of artificial grids - hours and days, longitude and latitude - on temporal and spacial landscapes, or perhaps I should say on timescapes and landscapes. Both provide a way of reversing divisibility, of retrieving unity, of recapturing a sense of the whole, even though it can never be the whole.... [For,] whether they take the form of crude markings in the sand, or of the most sophisticated computer graphics, maps have in common, as do the works of historians, a packaging of vicarious experience.... [And, perhaps most importantly,] there’s no such thing as a single correct map”
(Gaddis, pp.32-3)

“So, what if we were to think of history as a kind of mapping?... It would establish the linkage between pattern-recognition as the primary form of human perception and the fact that all history - even the most simple narrative - draws upon the recognition of such patterns. It would permit varying levels of detail, not just as a reflection of scale, but also of the information available at any given time.... But, most important, this metaphor would allow us to get closer to the way historians know when they’ve got it right. For verification in cartography takes place by fitting representations to reality. You have the physical landscape, but you wouldn’t want to try to replicate it. You have, in your mind, reasons for representing the landscape: you want to find your way through it without having to solely rely on your own immediate senses; hence you draw on the generalized experience of others. And you have the map itself, which results from fitting together what is actually there with what the user of the map needs to know about what is there.”
(Gaddis, pp.33-4)

As this suggests, Gaddis' key analogy - to which he returns repeatedly - is with mapmaking, and with the past as a landscape. The comparison is extremely fruitful, allowing fresh understandings of matters too easily glossed over. As he argues, there can no such thing as one perfect map - all maps are selective and designed to serve specific purposes. A geological map is of little use to the  motorist - although the reverse can be the case for geologists, who are often heavily dependent upon road cuttings for fresh exposures. The problem is inherently one of representation - since everything cannot be included, any representation is fundamentally both selective and a distortion of what it aims to portray. Rather than getting bogged down on this basic point, as Gaddis argues, we should evaluate histories in the same way as maps - by their usefulness. And such pragmatism, he insists, should properly be extended throughout a historian’s work:

It should also be noted that narrative predates history - and language itself, it seems - and needs to be "disciplined" in order to serve the historian's purposes. In consequence, the insights and assumptions which underlie the historian's craft are more scattered and opportunistic in nature than is the case with regard to other disciplines. They are also less counter-intuitive, as one might expect, and resemble the kind of checklist which could be constructed against sloppy thinking. Manipulations of time, space and scale are fundamental to narrative, but they are there to serve fundamentally pragmatic - rather than methodological - ends, and the test of their appropriate use is therefore pragmatic, one of fit with the questions posed in the enquiry, and the evidence brought to light. Still, as William McNeill has argued, given that historians’ practices have been better than their epistemology, perhaps theirs has been the wiser course...as opposed to the methodological obsessions so common in the social sciences.

Perhaps the core of Gaddis' argument in this book is a sustained comparison and contrast between historian's methods and those of the social and physical/biological sciences. In this, he repeatedly shows that, judged by a variety of criteria, the historian's approach is much closer to that of the historical sciences, such as geology and evolutionary biology, than it is to that of the social sciences, which have adhered much more closely to narrow nineteenth century ideas about "the" scientific method. And, although some of his strictures do not apply to much more sophisticated work such as Runciman's The Social Animal, the discussion is both apt, and telling.


“The remote sensing of processes by the way of surviving structures - whether in history or in science...is a deductive act: the task is to deduce the processes that produced it. You can hardly perform that task, though, without repeated acts of induction: you have to survey the evidence, sense what’s there, and find ways to represent it. Finding those ways, though, gets you back to the deductive level, for you must deduce them from the interests of those for whom the representation is being made. It makes little sense, then, to try to align structure and process neatly with deduction and induction. What’s required instead is to apply both techniques to the objects of your inquiry, fitting each to the other as seems appropriate to the task at hand.... Some years ago, I asked the great global historian William H. McNeill to explain his method of writing history to a group of social, physical, and biological scientists attending a conference I’d organized. He at first resisted doing this, claiming that he had no particular method. When pressed, though, he described it as follows:

        I get curious about a problem, and start reading up on it.  What I read causes me to redefine the
        problem. Redefining  the  problem  causes  me to shift the direction of what I'm reading. That in
        turn  further  reshapes  the problem,  which further redirects the reading.  I go back and forth like
        this until it feels right,  then I write it up and ship it off to the publisher.

McNeill’s presentation elicited expressions of disappointment, even derision, from the economists, sociologists, and political scientists present. ‘That’s not a method,’ several of them exclaimed. ‘It’s not parsimonious, it doesn’t distinguish between independent and dependent variables, it hopelessly confuses induction and deduction.’ But then there came a deep voice from the back of the room. ‘Yes, it is,’ it growled. ‘That’s exactly the way we do physics!’"
(Gaddis, pp.46-8)

One major difference lies in the uses of narrative. Social scientists embed narratives within generalizations, subordinating narrative to hypothesis testing. In direct contrast, historians generalize for specific narrative purposes - to explore what is general in unique events - so here, generalizations are subordinated to narrative. In line with this, the generalizations used are limited, and explanations are tied to both contingent and general forms of causation.

Akey distinction also lies in the historian's preference for simulations as opposed to modelling. This point is rarely made in the literature, as few appear to note that narrative can be seen as a form of simulation. The difference between these approaches is that modelling relies on stripped-down representations in which causal links are all clearly defined. Simulations, as Gaddis uses the term, make no such assumptions, and are attempts to illustrate past events through representation. While modelling is clearly the more powerful tool if used correctly, its limitations in the face of tangles of causation are also clear, and methodological modesty makes good sense when confronted by the proliferation of causes that characterize human history.

Causation itself provides a further marker of difference. Social scientists, emulating their experimental counterparts in the physical sciences, prefer parsimony in causes - the "independent variable" which makes experimentation a valuable procedure. Historians, in contrast, are happy to work with multiple - and interacting - causes, restricting their parsimony to consequences. This just basically means that you don't explain events by reference to distant and general factors, but try to keep the causal chains short - if necessarily tangled.

Historians also tend to mix inductive and deductive logics, use multiple explanatory frameworks, and probably don't chew their food properly - at least in the eyes of the more narrow-minded methodologists. From a broader perspective, however, things look considerably different. Recent work in the philosophy of science - based more closely on actual practice - acknowledges that induction and deduction necessarily mix in actual scientific work. And similarly, the recent scientific revival of William Whewell's concept of  consilience - for the unexpected coincidence of results drawn from different parts of a subject - underscores the value of using multiple paradigms. As Gaddis observes:


“I once had an article turned down by a major international relations journal on the grounds that I’d indulged in paradigm pluralism. ‘Not allowed,’ the reader’s report read. ‘You can only have one paradigm at a time.’ After brooding about this for a long time, I’ve concluded - hardly surprisingly - that that’s a short-sighted view. I’d cite as my authority, William Whewell, who argued a century ago that a situation of ‘rules springing from remote and unconnected quarters [but leaping] to the same point’ was possible only ‘from that being the point where truth resides.... [So,] historians are - or ought to be - open to diverse ways of organizing knowledge: our reliance on micro- rather than macro-generalization opens up for us a wide range of methodological approaches....We're free to describe, evoke, quantify, qualify, and even reify if these techniques serve to  improve the ‘fit’ we're trying to achieve. Whatever works, in short, we should use. Of course, it's pragmatic, inconsistent, and often just plain messy. But it is, I believe, good science, for what we can learn should always figure more prominently in our set  of  priorities than the purity of the methods by which we learn it.”
(Gaddis, pp.108-9)

It's also important to note that much of the common social science practice analyzed by Gaddis does not necessarily follow from the basic mindset identified by Runciman in The Social Animal. But again, perhaps this shouldn't be surprising. Herodotus, famed as the "father of history" is also frequently cited as the first anthropologist, and social science thinking has always been part of the historian's toolkit. Arguably, the tension between social scientists and historians has more to do with now discredited philosophies of science than it does with genuine conflict between the two approaches at a fundamental level.

Historians need the insights into structure that social science thinking provides, and social scientists need historical thinking if they are not to remain trapped at the level of static structures. There's no substitute - we need different modes of thought in order to make sense of things. But, we also need to adopt a more disinterested attitude towards them. Shorn of partisanship, valuing one mode over another in general is exactly like saying a hammer is better than a screwdriver - it literally doesn't make sense. My preference for historical approaches, I think, stems mostly from their natural human legibility - too often today we value the counter-intuitive simply because it's difficult, without stopping to ask how useful it is. But mine is a preference rooted in reception - it does not have anything to say about the applicability of different mindsets. As always, the test - as Gaddis repeatedly states - is pragmatic. What works is best.


“So, is history a science? I put the question to a group of Yale seniors recently, and the answer one of them came up with made perfect sense to me: it was that we should instead concentrate on determining which sciences are historical. The distinction would lie along the lines separating actual replicability as the standard for verification - the rerunning of experiments in a laboratory - from the virtual replicability that’s associated with thought experiments. And it would be the accessibility versus the inaccessibility of processes that would make the difference.”
(Gaddis, p.43)

John Lewis Gaddis’ The Landscape of History is the most engaging and wide-ranging methodological work I have ever read. Here, homo narrans finally gets its due from the camp of history, whilst the blinkered & prejudiced viewpoints re history all-too-common amongst social scientists are comprehensively demolished, and history’s real kinship - with certain of the “hard” sciences is clearly shown. Not only that, but we also have an insightful exploration of the historical mindset on offer...and, all in one hundred and fifty pages, what’s more.

And so, just what does it all add up to? Discipline your narrative by close attention to questions and evidence, keep your causal chains short but accept multiple interacting causes, and achieve consilience through judicious but selective use of multiple explanatory frameworks. It's probably the most basic form of educated common-sense, so we shouldn't be surprised that there's nothing counter-intuitive involved.


“What you hope for, as aresult of such teaching, is a present and future upon which the past rests gracefully.... I mean by this a society prepared to respect the past while holding it accountable, a society less given to uprooting than to retrofitting, a society that values moral sense over moral insensibility. Historical consciousness may not be the only way to build such a society; but just as, within the realm of nonreflexive entities, the scientific method has shown itself more capable than other modes of inquiry in commanding the widest possible consensus, so the historical method may occupy a similarly advantageous position when it comes to human affairs.”
(Gaddis, p.149)




John Henry Calvinist