A Brief History of Arrangement

This article is a “prequel” to the blog post in July on ”Building Information Neighbourhoods“. It tries to demythologise the undue influence that Linnaeus had on how we define the forms that taxonomies take. It’s also an extract from my forthcoming book.

A brief history of arrangement

In his well known book The Geography of Thought, Richard Nisbett argues that Western cultures tend to think in categories, while East Asian cultures tend to think in contexts. So if an Asian and a Westerner are asked to group associated things from a list comprising Monkey, Cow, Banana and Grass, the Asian will tend to connect Monkey with Banana and Cow with Grass, while the Westerner will tend to connect the animals with animals, and the foods with foods. Asians see connections and contexts, while Westerners see categories (Nisbett 2003).

Now Nisbett rather over-argues his case, and himself admits that Westerners can think in contexts and Asians can think in categories if they are primed to do so. But his distinction between categorical thinking and contextual thinking is a useful one to pursue. In the history of taxonomy work, this distinction turns out to be a critical one.

Until the 1730s, at least in Europe, the practices of arrangement, classification, categorization, and taxonomy work, were all part of the same cluster of concepts. As we saw in a previous post, the term taxonomy itself literally means habits or laws of arrangement. By the mid-1770s however, taxonomy work had already diverged from arrangement work, and this decisive split is exemplified by the conflict between two great biologists – George Louis Leclerc Comte de Buffon, and Carl Linnaeus.

Born in the same year (1707), Linnaeus and Buffon occupied very different social stations, sat at opposite ends of Europe, and set out opposing positions on the ordering and arrangement of knowledge about species. Linnaeus fought on the side of analysis, categories and controlled nomenclature, while Buffon fought on the side of context, multiple perspectives and understanding through seeing the bigger picture. Though Linnaeus won the field in terms of how taxonomy work became understood, Buffon had a profound though less direct impact on the way in which knowledge was organised for use in subsequent generations. He was the distant ancestor of the disciplines of ecology, information architecture and facetted classification.

To see how this played out, it’s necessary to go back in time a couple of centuries. Arrangement is an art that only becomes important when you have a collection of things to arrange. Throughout late antiquity and the middle ages, the few people who had the resources to collect diverse things in large quantities were nobles, clerics and royalty, and they did this for prestige, diversion and display. Arrangements of their “cabinets of curiosities” served to inspire wonder and impress their visitors.

Throughout the fifteenth century, with the spreading of wealth through trade and the growth of scholarship, there was a gradual shift in the purpose behind collection and arrangement. The passion for collection of “curiosities” was taken up on a large scale by scholars and scientists across Europe, and their collections were increasingly used as instruments of learning about the natural world. Arrangements of curiosities became part of a larger endeavour to construct a systematic knowledge of the natural world. Collections started to become more systematic and supportive of enquiry, sensemaking and discovery. By the beginning of the seventeenth century, writers like Francis Bacon were thoroughly dismissive of the higgledepiggledy arrangements of the rich and famous:

“There is such a multitude and host as it were of particular objects, and lying so widely dispersed, as to distract and confuse the understanding; and we can therefore hope for no advantage … unless we put its forces in due order and array by means of proper, and well arranged, and as it were living tables of discovery of these matters which are the subject of investigation…” (Blom 2003:46).

As knowledge of the world grew with the expansion of exploration, trade, imperial and economic dominion throughout the seventeenth century, so the collections of the curious grew, and the challenges of sensible arrangement became even more pressing. Bacon’s impatience was echoed just over a century later by the methodical Carl Linnaeus who was dismissive of the “complete disorder” he found in the home of the last great universal collector of his time, Sir Hans Sloane – founder of the collection that became the British Museum. After Sloane, collectors divided themselves into discrete disciplines. The world of knowledge had become too complex to comprehend and represent in one single arrangement (Blom 2003:88).

In the midst of this complexity, Linnaeus’ great gift to science was threefold. Beginning with his Systema Natura in 1735, he introduced a far simpler principle of distinguishing between species based on anatomy than had ever been proposed before. Beginning in 1737 with his Critica Botanica he laid down the rules for his binomial naming system for species which riled his critics immensely (because he substituted so many older naming conventions with his own), but when widely adopted created the first standardized way of describing species. This immeasurably enhanced scientific coordination and collaboration. Finally, his hierarchical, nested classification tree structure turned out to be a perfect vehicle to express the genealogical relationships that gained such prominence during the emerging evolutionary theories of the late eighteenth and early nineteenth centuries (Blunt 2001; Gould 2000:80).

Linnaeus’ new taxonomic method simplified the task of categorization, imposed rigourous rules (and therefore consistency), and happened on a form of representation that history turned into a lucky bet. From the point of view of advancing scientific method, his focus on analysis, rules and standardised approaches, gave an incalculable advantage.

Buffon by comparison arrived late on the biological scene, only taking up his position of director of the king’s botanic gardens in Paris in 1738, when Linnaeus’ work on botany was already beginning to reverberate through Europe. He had been a brilliant mathematician, and had at the age of twenty invented the binomial theorem in mathematics (related only in name to Linnaeus’ binomial system of nomenclature). But biology turned out to be Buffon’s lifelong passion. Beginning in 1749, he began an astonishing forty year odyssey of publications under the title Histoire Naturelle, Générale et Particulière (Natural History: General and Particular). Projected to run to 50 volumes, Buffon lived to see 36 of them published, and the series is cited by Stephen Jay Gould as “one of the most comprehensive and monumental efforts ever made by one man” (Gould 2000:78).

Buffon vehemently opposed what he saw as Linnaeus’ unhelpful over-simplifications. Buffon recognized the inherent complexity of biological life, and insisted that this complexity be represented faithfully. There were many different ways of relating organisms to each other, based on different attributes and principles. They could be grouped according to the environments that they shared (eg air, land or water). They could be grouped by function and adaptation (eg having wings and flying brings bats closer to birds than mammals, whereas anatomy separates the two). They could be grouped by similarity of behaviour, and so on. Anatomy provided one, but only one way of classifying creatures. The only true test of a species was whether inter-breeding between animals was possible so genealogical lineage was also a grouping criterion, leading to Buffon’s insight into the instability of species, and their adaptations over time (Gould 2000:75-90). Buffon was implicitly adopting a facetted approach to the organization of species.

Linnaeus, as we know, won the war, and it’s not hard to understand why. Simplicity and standardisation paid off for the science of the time much more than recognition of complexity, context, and multiple ways of looking at things. If ecology had become the mantra of the following century rather than evolution, then Buffon’s attention to context, environment and adaptation might have brought his theory back to prominence – indeed, Buffon is now recognised by ecological thinkers as an early fore-runner. But it was not to be. In 1774 Buffon had to swallow the Linnaean system when king Louis XV decreed that it be adopted in the king’s gardens.

But if Linnaeus successfully appropriated and narrowed the discipline of scientific taxonomy, Buffon’s legacy lived on in the arrangements of things for understanding, sensemaking and education. Museums, galleries and universities adopted the “Buffonian system” with enthusiasm, because of its instructive power. The founder of America’s first museum, Charles Willson Peale, presented his exhibits with painted backdrops and artificial landscapes, lurking with their environmental neighbours. At the end of the eighteenth century, the Emperor Franz II constructed a museum in Vienna which was even more lavish than Peale’s in its reconstructions of natural habitats, though it sometimes sacrificed accuracy for entertainment.

Also in Vienna in the 1780s, the Hapsburg picture collection was reorganized by Christian van Mechel into chronological order, and suddenly opened up the possibility of art history – identifying styles, schools, and periods in art. This vision was brought to fruition in post-revolutionary France right at the end of the eighteenth century when Dominique Denon reorganized the collections of the Louvre on historical principles (Blom 2002).

Buffon’s great gift then, was in the recognition that there are many possible ways to organize the same things, and that every arrangement tells a different story. This is what brings us to information architecture.

The inventor of the term ‘information architecture’ is Richard Saul Wurman, a ‘real’ architect by training and avocation, but who has increasingly turned to the design problems involved in providing access to information. Wurman points out: “The ways of organizing information are finite. It can only be organized by location, alphabet, time, category, or hierarchy” (Wurman 2001:40-41).

But what Wurman also points out is that the same set of things can often be arranged in each of those ways, and each arrangement tells you different things. If you consider an arrangement of pedigree dogs, for example, an arrangement by size displays its own patterns and suggests its own questions (like “why are there so few large pedigree dogs compare with tiny ones?” ). An arrangement by country of origin tells another story, while an arrangement by date of Kennel Club recognition tells a tale of changing tastes over time.

The task of the information architect therefore is to find the arrangements that will be most instructive and useful for any given context of use. Arrangements, not classifications, are the primary name of the game.

This is what the taxonomist, even the facetted taxonomist, is apt to forget if overly focused on the analytical task of discriminating and describing according to fixed principles. Too much attention to the category infrastructure represented by your taxonomy can obscure proper attention to the contexts of use for the information and knowledge assets you cover, and this is why your relationship with the information architects is so important.

And this is why I think the attention of the taxonomist needs to extend to “Building Information Neighbourhoods”!

4 Comments so far

Wow, your explanation of classification vs. arrangements is very helpful. I’m ordering Geography of Thought tonight.

I think the concepts in the article can be applied to understanding why some software manuals are so awful. I recently helped a family member try to find a good manual for Access. Plenty of books had pretty screenshots and looked promising at first glance. At second glance, I realized many of the procedures were not connected and lacked context.

My experiences with poorly designed manuals can be compared with trying to learn how to drive a car by looking up “tire”, “mirror”, “transmission”, etc… in unrelated encyclopedia articles.

Your article and the one by Maish Nichani (Taming your “target” Content) provide the framework with which to explain information confusion.

Posted on September 25, 2006 at 08:52 AM | Comment permalink

Patrick Lambe

I like your example of the manuals, Larry (may I borrow that example in future?), and I’m delighted you found the article useful.

Posted on September 25, 2006 at 01:33 PM | Comment permalink

Godfried Augenbroe

I enjoyed this article enormously.
I am so looking forward to the book, which btw is still not available on the US amazon site..??!!

One minor comment: Although writing Hapsburg instead of Habsburg is done frequently in English literature, most academics have adopted Habsburg as the correct spelling.

Posted on January 17, 2007 at 01:26 AM | Comment permalink


Many thanks Godfried, for the Habsburg pointer and the anticipation for the book! The book should be out any day now, and Im told it will then appear on Amazon US - they don’t do pre-ordering for books from non-US publishers, perhaps? It’s on the Amazon.co.uk site but that’s not much help if you’re in the US.

Do let me have any feedback you have when you finally get the book.

Posted on January 18, 2007 at 02:25 AM | Comment permalink

Page 1 of 1 pages

Commenting is not available in this weblog entry.

Comment Guidelines: Basic XHTML is allowed (<strong>, <em>, <a>) Line breaks and paragraphs are automatically generated. URLs are automatically converted into links.