-------------------------------------------------------------------------------
Return to article page

To print: Select File and then Print from your browser's menu.
-------------------------------------------------------------------------------
This story was printed from FindArticles.com, located at http://www.findarticles.com.
-------------------------------------------------------------------------------


Natural History
Nov, 1998

Writing in the margins.

Author/s: Stephen Jay Gould

Famed as the greatest of chemists, Lavoisier made seminal contributions to geology that have gone unheralded.

I once had a teacher with an idiosyncratic habit that distressed me forty years ago but now and finally, oh sweet revenge--can work for me to symbolize the general process of human creativity. I never knew a stingier woman, and though she taught history in a New York City junior high school, she might well have been the frugal New England farmer with the box marked "pieces of string not worth saving." Readers who attended New York City public schools in the early 1950s will remember those small yellow slips of paper, three by six inches at most, that served all purposes, from spot quizzes to "canvases" for art class. Well, Mrs. Z. would give us one sheet--only one--for any classroom exam, no matter how elaborate the required answers. She would always reply to any plea for advice about containment or, God forbid, an additional yellow sheet (comparable in her system of values to Oliver Twist's request for more soup) with a firm refusal followed by a cheery instruction expressed in her oddly lilting voice: "And if you run out of room, just write in the margins!"

Margins play an interesting role in the history of scholarship, primarily for their schizophrenic housing of the two most contradictory forms of intellectual activity. Secondary commentaries upon printed texts (often followed by several layers of commentaries upon the commentaries) received their official designation as "marginalia" to note their necessary position at the edges. The usual status of such discourse as derivative and trivial, stating more and more about less and less at each iteration, leads to the dictionary definition of marginalia as "nonessential items" (Webster's Third New International Dictionary) and inevitably recalls the famous lines of Jonathan Swift:

So, naturalists observe, a flea

Hath smaller fleas that on him prey;

And these have smaller still to bite 'em;

And so proceed ad infinitum.

Thus every poet, in Iris kind,

Is bit by him that comes behind.

But margins also serve the diametrically opposite purpose of receiving the first fruits and inklings of novel insights and radical revisions. When received wisdom has hogged all the central locations, where else can creative change begin? The curmudgeon and cynic in me regards Thoreau's Walden as the most overquoted (and underwhelming) American classic, but I happily succumb, for the first time, to cite his one-liner for a vibrant existence: "I love a broad margin to my life."

Literal margins, however, must usually be narrow--and some of the greatest insights in the history of human thought necessarily began in such ferociously cramped quarters. The famous story of Fermat's last theorem, no matter how familiar, cannot be resisted in this context: When the great mathematician Pierre Fermat died in 1665, his executors found the following comment in his copy of Diophantus' Arithmetica, next to a discussion of the claim that no natural numbers x, y, and z exist such that [x.sup.n] + [y.sup.n] = [z.sup.n], where n is a natural number greater than 2: "I have discovered a truly remarkable proof but this margin is too small to contain it." Mathematicians finally proved Fermat's last theorem just a few years ago, to great subsequent fanfare and an outpouring of popular books. But we shall newer know if Fermat truly beat the best of the latest by 350 years or if (as my own betting money says, admittedly with no good evidence) he had a promising idea but never saw its disabling flaw in the midst of his excitement.

I devote this essay to the happier and opposite story of a great insight that a cramped margin did manage Oust barely) to contain and nurture, but which then grew to such originality and fruitfulness that I need a two-part essay to do the subject justice. Part I appears here, and Part 2 will be published in next month's issue. But this tale, for reasons that I do not fully understand, remains virtually unknown (and marginal in this frustrating sense) both to scientists and to historians alike, although the protagonist ranks as one of the half-dozen greatest scientists in Western history, and the subject stood at the forefront of innovation in his time. In any case, the movement of this insight from marginality in 1760 to centrality by 1810 marks the birth of modern geology and gives us a rare and precious opportunity to eavesdrop on a preeminent thinker operating in the most exciting and instructive of all times: at the labile beginning of the codification of a major piece of natural knowledge --a unique moment featuring a landscape crossed by a hundred roads, each running in the right general direction toward a genuine truth. Each road, however, reaches a slightly different Rome and our eventual reading of nature depends crucially upon the initial accidents and contingencies specifying the path actually taken.

In 1700, all major Western scholars believed that the earth had been created just a few thousand years earlier. By 1800, nearly all scientists accepted a great antiquity of unknown duration and a sequential history expressed in the strata of the earth's crust. These strata, roughly speaking, form a vertical pile, with the oldest layers on the bottom and the youngest on top. By mapping patterns of the exposure of these layers on the earth's surface, this sequential history can be inferred. By 1820, detailed geological maps had been published for parts of England and France, and general patterns had been established for the entirety of both nations. This discovery of "deep time," and the subsequent resolution of historical sequences by geological mapping, must be ranked among the sweetest triumphs of human understanding.

Few readers will recognize the name of Jean-Etienne Guettard (1715-86), a leading botanist and geologist of his time and the instigator of the first "official" attempt to produce geological maps of an entire nation. In 1746, Guettard presented a preliminary "mineralogical map" of France to the Academic des Sciences. In subsequent years, he published similar maps of other regions, including parts of North America. As a result, in 1766, the secretary of state in charge of mining commissioned Guettard to conduct a geological survey and to publish maps for all of France. The projected atlas would have included 230 maps, but everyone understood, I suspect, that such a task must be compared to the building of a medieval cathedral and that no single career or lifetime could have completed the job. In 1770, Guettard published the first 16 maps. The project then became engulfed by political intrigue and, finally, by a revolution, which (to say the least) tended to focus attention elsewhere. Only 45 of the 230 projected maps ever saw the published light of day, and control of the survey had passed to Guettard's opponents by this time.

Guettard's productions do not qualify as geological maps in the modern sense, for he made no effort to depict strata or to interpret them as layers deposited in a temporal sequence--the revolutionary concepts that validated deep time and established the order of history. Rather, as his major cartographic device, Guettard established symbols for distinctive mineral deposits, rock types, and fossils--and then merely placed these symbols over appropriate locations on his map. We cannot even be sure that Guettard understood the principle of superposition; the key concept that time lies revealed in a vertical layering of strata, with younger layers above (superposed upon) older beds. Guettard did develop a concept of bandes, or roughly concentric zones of similar rocks, and he probably understood that a vertical sequence of strata might be expressed as such horizonal zones on a standard geographical map. But, in any case, he purposely omitted these bandes on his maps, arguing that he only wished to depict facts and avoid theories.

This focus on each factual tree, combined with his studious avoidance of any theoretical forest of generality or explanation, marked Guettard's limited philosophy of science, and also (however unfairly) restricted his future reputation, for no one could associate his name with any advance in general understanding. Rhoda Rappoport, a distinguished historian of science at Vassar College and the world's expert on late-eighteenth-century French geology, writes of Guettard (within a context of general admiration, not denigration): "The talent he most conspicuously lacked was that of generalization, or seeing the implications of his own observations .... Most of his work reveals... that he tried hard to avoid thinking of the earth as having a history."

But if Guettard lacked this kind of intellectual flair, he certainly showed optimal judgment in choosing a younger partner and collaborator for his geological mapping, for Guettard fully shared this great enterprise with Antoine-Laurent Lavoisier (1743-94), a mere fledgling of promise at the outset of their work in 1766 and the greatest chemist in human history when the guillotine cut short his career in 1794.

Guettard and Lavoisier made several field trips together, including a four-month journey in 1767 through eastern France and part of Switzerland. After the first sixteen maps were completed in 1770, Lavoisier's interest shifted away from geology toward the sources of his enduring fame--a change made all the more irrevocable in 1777, when control of the geological survey passed to Antoine Monnet, inspector general of mines and Lavoisier's enemy. (Later editions of the maps ignore Lavoisier's contributions and often don't even mention his name.)

Nonetheless, Lavoisier's geological interests persisted, buttressed from time to time by a transient hope that he might regain control of the survey. In 1789, with his nation on the verge of revolution, Lavoisier published his only major geological paper--a stunning and remarkable work that shall occupy the second installment of this essay. Amid his new duties as regisseur des poudres (director of gunpowder) and leading light of the commission that invented the meter as a new standard of measurement--and despite the increasing troubles that would lead to his arrest and execution (for his former role as a farmer general, or commissioned tax collector)--Lavoisier continued to write notes about his intention to pursue further geological studies and to publish his old results. But the most irrevocable of all changes fractured these plans on May 8, 1794, less than three months before the fall of Robespierre and the end of the Terror. The great mathematician Joseph-Louis Lagrange lamented the tragic fate of his dear friend by invoking the primary geological theme of contrasting time scales: "It took them only an instant to cut off his head, but France may not produce another like it in a century."

All the usual contrasts apply to the team of Guettard and Lavoisier: established conservative and radical beginner; mature professional and youthful enthusiast; meticulous tabulator and brilliant theorist; counter of trees and architect of forests. Lavoisier realized that geological maps could depict far more than the mere location of ores and quarries. He sensed the ferment accompanying the birth of a new science, and he understood that the earth had a long history, potentially revealed in the rocks on his maps. In 1749, Georges Buffon, the greatest of French naturalists, had begun his monumental treatise--Histoire naturelle--which would eventually run to forty-four volumes, with a long discourse on the history and theory of the earth.

As Lavoisier groped for a way to understand this history from the evidence of his field trips, and as he struggled to join the insights published by others with his own original observations, Lavoisier recognized that the principle of superposition could yield the required key: the vertical sequence of layered strata must record both time and the order of history. But vertical sequences differed in all conceivable ways from place to place--in thickness, in rock types, in the order of the layers. How could one take--this confusing welter and infer a coherent history for a large region? Lavoisier appreciated the wisdom of his older colleague enough to know that he must first find a way to record and compile the facts of this variation before he could hope to present any general theory to organize his data.

Lavoisier therefore suggested that a drawing of the vertical sequence of sediments be included alongside the conventional maps festooned with Guettard's symbols. But where could the vertical sections be placed? In the margins, of course--for no other space existed in the completed and conventional design. Each sheet of Guettard and Lavoisier's atlas therefore features a large map in the center with a nnarginal column on either side: a tabular key for Guettard's symbols m the left margin and Lavoisier's vertical sections in the right margin. If I wished to epitomize the birth of modern geology in a single phrase (admittedly oversimplified, as all such efforts must be), I would honor the passage--both conceptual and geometric--of Lavoisier's view of history, as revealed in sequences of strata, from a crowded margin to the central stage.

Many fundamental items in our shared conceptual world seem, obvious and incontrovertible only because we learned them in our cradle (so to speak) and have never even considered that alternatives might exist. We often regard such notions --including the antiquity of the earth, the rise of mountains, and the deposition of sediments--as simple facts of observation, so plain to anyone with eyes to see that any other reading could arise only from the province of knaves or fools. But many of these "obvious" foundations emerge as difficult and initially paradoxical conclusions born of long struggles to think and see in new ways.

If we can recapture the excitement of such innovation by temporarily suppressing our legitimate current certainties and reentering the confusing transitional world of our intellectual forebears, then we can understand why all fundamental scientific innovation must marry new ways of thinking with better styles of seeing. Neither abstract theorizing nor meticulous observation can provoke a change of such magnitude all by itself. And when--as in this story of Lavoisier and the birth of geological mapping--we can link one of the greatest conceptual changes in the history of science with one of the most brilliant men who ever graced the profession, then we can only rejoice in the enlarged insight promised by such a rare, conjunction.

Most of us, with minimal training, can easily learn to read the geological history of a region by studying the distribution of rock layers on an ordinary geographical map and then coordinating this information with vertical sections (as drawn in Lavoisier's margins) representing the sequence of strata that would be exposed by digging a deep hole in any one spot. But consider, for a moment, the intellectual stretching thus required, and the difficulty that such an effort would entail, if we didn't already understand that mountains rise and erode and that seas move in and out over any given region of our ancient earth.

A map is a two-dimensional representation of a surface; a vertical section is a one-dimensional listing along a line drawn perpendicular to this surface and into the earth. To understand the history of a region, we must mentally integrate these two schemes into a three-dimensional understanding of time (expressed as vertical sequences of strata) across space (expressed as horizontal exposures of the same strata on the earth's surface). Such increases in dimensionality rank among the most difficult of intellectual problems--as anyone will grasp by reading the most instructive work of science fiction ever published, Edwin A. Abbott's Flatland (1884, and still in print), a "romance" (his description) about the difficulties experienced by creatures wino live in a two-dimensional world when a sphere enters the plane of their entire existence and forces them to confront the third dimension.

As for the second component of our linkage, I can offer only a personal testimony, My knowledge of chemistry is rudimentary at best, and I can therefore claim no deep understanding of Lavoisier's greatest technical achievements. But I have read several of his works and have never failed to experience one of the rarest emotions in my own arsenal: sheer awe accompanied by spinal shivers. A kind of eerie, pellucid clarity pervades Lavoisier's writing (and simply makes me ashamed of the peregrinations in these essays).

Perhaps, indeed almost certainly, a few other scientists have combined equal brilliance with comparable achievement, but no one can touch Lavoisier in shining the light of logic into the most twisted corners of old conceptual prisons and onto the most tangled masses of confusing observations, and extracting new truths expressed as linear arguments accessible to anyone. As an example of the experimental method in science (including the fundamental principle of double-blind testing), no one has ever bettered the document that Lavoisier wrote in 1784 as head of a royal commission (including Benjamin Franklin, then resident in Paris, and, ironically, Dr. Guillotin, whose "humane" invention would end Lavoisier's life) to investigate (and, as results proved, to refute) the claims of Franz Mesmer about the role of "animal magnetism" in the cure of disease by entrancement (mesmerization).

Lavoisier did not publish his single geological paper (analyzed in next month's column) until 1789, but Rhoda Rappoport has shown that he based this work upon conclusions reached during his mapping days with Guettard. Lavoisier did not invent the concept of vertical sections, nor did he originate the idea that sequences of strata record the history of regions on an earth of considerable antiquity. Instead, he resolved an issue that may seem small by comparison but couldn't be more fundamental to any hope for a workable science of geology (as opposed to the simpler pleasures of speculating about the history of the earth from an armchair): he showed how the geological history of a region can be read from variations in strata from place to place--in other words, how a set of one-dimensional lists of layered strata at single places could be integrated by that greatest of all scientific machines, the human mind, into a three-dimensional understanding of the history of geological changes across an entire region.

(I doubt that Lavoisier's work had much actual influence, for he published only this one paper on the subject and did not live to realize his more extensive projects. Other investigators soon reached similar conclusions, for the nascent science of geology was the hottest intellectual property in late-eighteenth--century science. Lavoisier's paper has therefore been forgotten, despite several efforts by isolated historians of science through the years--with this two-part essay as the latest attempt--to show the singularity of Lavoisier's vision and accomplishment.)

From my excellent sample of voluminous correspondence from lay readers during a quarter century of writing these essays, I have learned the irony of the most fundamental misunderstanding about science among those who love the enterprise (I am not discussing the different errors made by opponents of science). Supporters assume that the greatness and importance of a work correlates directly with its stated breadth of achievement: minor papers solve local issues, while great works claim to fathom the general and universal nature of things. But all practicing scientists know in their bones that successful studies require strict limitations. One must specify a particular problem with an accessible solution, and then find a sufficiently simple situation where attainable facts might point to a clear conclusion. Potential greatness then arises from cascading implications toward testable generalities. You don't reach the generality by direct assault without proper tools. One might as well dream about climbing Mount Everest wearing a T-shirt and tennis shoes and carrying a backpack containing only an apple and a bottle of water.

I shall, in next month's essay, show how Lavoisier's only geological paper rests upon such a simple and testable theme--the claim that sea level rises and falls in cycles and that distinctive differences m type of sediment mark the lowstands and highstands. The argument of this remarkable work then cascades in two directions from this firm starting point: first, toward a great statement about the methodology of science and, subsequently, toward a framework for the full panoply of concepts that would elevate the diffuse and tentative efforts of naturalists into an expansive and cohesive science of geology.

Stephen Jay Gould teaches biology4 geology, and the history of science at Harvard University. He is also the Frederick P. Rose Honorary Curator in Invertebrates at the American Museum of Natural History.

-------------------------------------------------------------------------------
COPYRIGHT 1998 American Museum of Natural History
in association with The Gale Group and LookSmart. COPYRIGHT 2000 Gale Group
-------------------------------------------------------------------------------