Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Introduction

Ever since at least the days of Herbert Spencer, notions of “cultural evolution” have dotted the intellectual landscape. And if cultural evolution has not dominated cultural anthropology and archaeology to the degree to which it may be said to have dominated biology (and even there the impact and relevance of evolutionary ideas has been more checkered than is usually conceded), nonetheless the legitimacy and importance of notions of evolution in human cultural history have by and large seemed obvious to most observers (save perhaps British structural anthropologists).

Much of the work in cultural evolution in recent decades has tended to mirror—and often downright mimic—the approaches of hard-core neo-Darwinists in biological evolutionary theory. Boyd and Richerson’s (1985) influential volume, for example, comes to mind in this regard—as of course does Dawkins’ (1976) notion of “memes”—his parallel construct to his perhaps more elemental notion of the “selfish gene.” A strong, reductive selectionism—a sort of “one size fits all”—is the prevailing core concept in such endeavors. Nor is the paradigm mere analogy—as it is not uncommon to find notions of “fitness” incorporated into discussions of cultural evolution, even to the point of allegations that cultural evolutionary change enhances the “fitness” of the developers/exponents of cultural novelty. Build a stronger bow, and make more babies! This extreme line of thinking puts cultural evolution in the invidious position of being a form of biological evolution after all. Talk about reductionism!

This present volume, along with another recent effort by archeologists (Lipo, et al., 2006), by and large takes the ontologically far more compelling and realistic position that cultural systems—be they material cultural or social systems—are more complicated than a single “population,” where selection winnows each generation’s standing crop of variation. Selection works in such domains—but the larger-scaled systems and patterns that archeologists (like their biological counterparts—paleontologists) typically confront present degrees of complexity that prevent (or at least should prevent) the straightforward application of simple, linear selection models. Simple selection models will not work for entire species divided into semiautonomous demes (the picture of species structure hinted at in Darwin and developed explicitly by Sewall Wright). I doubt they work any better for complex social systems and will demonstrate below that they certainly are not appropriate for complex material cultural systems, such as those typical of manufactured goods in industrialized nations1.

It is not ipso facto wrong to seek parallels between biological and material cultural evolution in the attempt to clarify the underlying ontological structure and causalities within each system. Reciprocal illumination can work—especially if the underlying intent is not to derive—or simply import—a theory of evolutionary process for one system directly from the other. Before such comparisons can be made, however, it is helpful to start with a definition of “evolution” that is both suitable and appropriate to both such systems.

What is Evolution?

“Change through time” vaguely fits the bill for a general definition of evolution—though of course stability of systems (in some contexts known as “stasis”) is not directly embraced. In general this definitional phrase is too weak to be of much service.

I prefer “the long-term fate of transmissible information” as a definition of evolution in both biological and material cultural systems. Stronger yet would be the more restrictive “long-term fate of transmissible information in an economic context.” The objection to adding economics to the definition is simply that both genetic and material cultural information can be added or lost or modified strictly through the processes of transmission themselves—as when sexual selection modifies reproductive features of organisms for reasons having nothing to do with organismic relative economic success; or perhaps when information about making stone tools is modified as a simple by-product of errors in transmission of the original information (as in the parlor game when a message is invariably changed when serially whispered into enough ears).

Another objection to adding “economics” to the definition is that it is possible simply to chart the actual historical fate of information in both systems: This is what “phylogenetics” is all about in biology—with some schools of thought notorious for their aversion to considerations of underlying causal theory. A cladogram will suffice—yet cladograms are undeniably approximations of actual evolutionary history. Such is also the case for descriptions of the histories of material cultural systems: spear point design—or the design history of cornets—can be described in and of itself, with no further analysis of the causal factors underlying that history.

On the other hand, most evolutionary biologists seek to achieve an enhanced understanding of underlying evolutionary process based upon—and transcending—their phylogenetic diagrams. And such considerations almost automatically necessitate a consideration of the economic (“ecologic”) context in which organisms lead their lives. Likewise, I really cannot think of material culture manufacture without considering the economic context of the objects being manufactured.

At the very least, the vast majority of long-term stasis and change in genetic information pertains to the economic side of organism’s lives: the conformation of the soma, the adaptations for the basic physiological functions like respiration, digestion, etc. Darwin’s classic formulation of natural selection boiled down to the effects that relative economic success (among conspecific organisms in a population) have on the relative reproductive success of those same organisms—given the competition for resources that limit the ultimate size of populations. And, of course, most things made by humans serve some sort of purpose, playing more-or-less obvious roles in the economics of daily life.

Returning to “fitness” for a moment, we can ask if there is an exact parallel between relative fitness—the statistical correlation between reproductive and economic success in biological populations2—in the biological realm, with a similar phenomenon in material cultural evolution. I have already commented that equating stasis and change in material cultural systems with patterns of relative reproductive success of humans (and even groups of humans) involved with designing, manufacturing, and using such products is tenuous at best.

The obvious difficulty of linking human reproductive success with the relative “success” of items of material culture in the marketplace highlights what at first glance at least is the beauty of Dawkins’ concept of “memes.” As with the parallel (and antecedent, really) notion of the “selfish gene,” what’s really going on in both biological and cultural systems is that it is the unit of information—the gene, the meme—whose “reproductive success” is really at stake. Were this really the case, we would have no difficulty in agreeing with Dawkins and those who see memes as the underlying basis for understanding cultural evolution.

But is it rational to see evolution in either system as a matter of bits of information competing with one another to be relatively better represented in the next succeeding generation? Despite Dawkins’ (1982) own distinction between the information in a system, on the one hand and the “vehicles” that carry the information on the other, to follow Dawkins, one has to suppose that it is the bits of information themselves that are really the active competitors for enhanced representation as time goes by.

Such teleology fails because it is impossible to imagine how a piece of information—be it ensconced in a genome, a patent drawing, or in an actual spear point accompanied with an oral tradition specifying its manufacture—could itself be in competition with alternative forms of the “same” information. Squirrels may compete for acorns, and people may compete for recognition as the greatest cornet player on earth—but the information underlying squirrel behavior and anatomy is only a passive player in the game. And the information underlying various different cornet models—which may well play a role in the contest between their human players to see who is indeed the best—is likewise confined to a passive response to various forms of competition. Cornet models grow and diminish in popularity through time, but it is conceptually vacuous to describe such waxings and wanings in the borrowed jargon of genetics.

Thus there is no proper analogue of biological concepts of “fitness” in the material cultural realm—both because of the exiguous linkage between manufactured objects and human reproductive success, and because reducing natural selection to the rubric of the selfish gene, and producing a parallel—the “selfish meme”—into considerations of cultural evolution, forces information to play active competitive roles instead of the passive, driven-by-the-fates roles it actually plays in both systems. Think of it this way: Information doesn’t give a damn whether it exists at all, let alone whether it will still be around in greater or diminished frequencies as time goes by. As I have found in my own database on the history of the cornet, the reasons why cornet designs (whether in their entirety, or in piecemeal, “meme” fashion) become more popular sometimes, or die out in others, are sometimes a reflection of relative efficiency (“selection” to be sure—but without the baggage of “fitness”; sometimes it is a matter of building a better mousetrap); but it just as often seems to be a matter of esthetics of a fickle buying public—who may be following the voguish dictates of professional players arbitrarily choosing among a spectrum of functionally equally good designs. I have found that some makers made better instruments than others—even if the designs were indistinguishably the same—and even when the makers of the better instruments themselves switched designs, or added alternative designs to their catalogues. A surprisingly high percentage of details of cornet design history cannot be attributed to actual improvements in the “playability” of the instruments.

Information in Biological and Material Cultural Systems

It has long been appreciated that the modes of transmission of genetic and human cultural information are very different. Genes are evanescent corporeal bodies; the information encoded on them is variably transcribed or copied before the actual gene dies. There are no such analogues in the cultural realm—“memes” being the information associated with, or underlying, a specifiable piece of cultural behavior—or, more concretely, an attribute of a physical artifact. Thus the “bore size” of a cornet (i.e. diameter of the air passage measured at the valves) might indeed usefully be construed as a “meme.” But bore size is an abstract category of a cornet’s property—and the bore size of any given cornet is merely an expression of the underlying information, whether from a shop drawing, or a physical mold or mandrel. Memes are the information lying in the “memotype”—and knowledge of them comes strictly from the phenotype. Genes, however, are themselves physical entities and can be studied as such in their own right.

And it is of course widely appreciated that, with the exception of bacterial systems and (some) plant taxa, genetic transmission is entirely vertical—i.e., from parent(s) to offspring. Not so with the transmission of human cultural information—neatly exemplified by the simple act of reading this sentence. Most considerations of the differences in how information is spread in biological (genetic) systems and human cultural systems emphasize the potential for far greater speed in the latter—on occasion engendering further thoughts of potentially faster rates of evolution.3

But there is another direct consequence of the disparities in mode of information transmission: For the evolutionary histories of complex material cultural systems can be notoriously difficult to analyze. Using a portion of my cornet data, Tëmkin and Eldredge (2007) showed that the memetic information—the 17 variables used to describe cornet “morphology” in my data base—is nearly entirely “unlinked” in cornet history, meaning that makers were free to mix and match parts. Openly stealing from another’s designs, virtually all permutations and combinations were possible—including “retrofitting” older designs with new ideas. Bottom line: Information can easily spread back across separate branches (lineages) after they are already established as phylogenetically distinct. That is the equivalent, e.g., of innovations in dogs (Canidae) being able to spread back over to cats (Felidae).

That fact alone makes the straightforward application to cultural systems of the sort of phylogenetic reconstruction algorithms that are routinely applied to biological systems virtually impossible: The results are almost invariably gibberish. Tempering the raw computer output with what can be deduced about cornet history on other criteria (historical sequence of innovations in cornet morphology, patents, advertising, etc.) yields the sort of results in Fig. 11.1.

Fig. 11.1
figure 12_1_185909_1_En

Evolution of cornets. The vertical bars correspond to periods of manufacture of particular models (M). Shaded left and right areas correspond to instruments equipped with Stölzel and Périnet valves, respectively. Curved lines represent reticulations. Triangles mark several key innovations in cornet design: (1) valve number, (2) shifting of the second valve slide and valve alignment, (3) changing of bell exit position and bell placement, and (4) alteration of bell shape (“trumpetization”). From Tëmkin and Eldredge (2007), Fig. 11.2; for further information on analytic procedures underlying the generation of this diagram, see the caption for Tëmkin and Eldredge (2007), Fig. 11.2. Downloadable data files are available via the online version of this paper

Full size image
Fig. 11.2
figure 12_2_185909_1_En

Phylogeny of the Baltic psaltery. Shaded branches of the cladogram represent linguistic groups. Bootstrap values shown above branches (2000 replicates). Baltic group: Lat, Latvian; Lit, Lithuanian; Ural-Altaic group: Est; Estonian; Fin, Finnish (including Karelian); Set, Setu; Vep, Vepsian; Vot, Votic; Slavic group: Rus, Russian. For further details on analytic procedures, see Tëmkin and Eldredge (2007), Fig. 11.1 (caption)—from which this figure is derived

Full size image

One mitigating fact is that in relatively simpler systems—such as Tëmkin’s data on the Baltic psaltery (including some archaeological specimens!) or for that matter relatively simple lithic industries confined to single area cultural traditions—fare better when it comes to their analysis through biological phylogenetic techniques (Fig. 11.2). Such histories are perforce more linear—with innovations added seriatim, relatively few “branch points”—and thus a minimum of exchange of information across the branches of the tree.

As if this were not enough to establish the inherently greater complexity of the evolutionary histories of material cultural systems vs. biological evolutionary systems, consider the “Hannah Principle” (Eldredge 2006; Tëmkin and Eldredge 2007). The cornerstone assumption of phylogenetic analysis in biological systems is that evolution proceeds in greatest measure by the successive transformation of homologous traits4—meaning the “same” feature undergoes different forms of transformation in collateral lineages, and the sequence of such transformations can be resurrected through phylogenetic analysis. Of course, in deep phylogenetic history, entirely novel features appear (though even these ultimately have a progenitor). Thus a straightforward goal of phylogenetic analysis is to distinguish homologous similarities (synapomorphies) from the false signals of nonhomologous adventitious similarity (e.g., through convergent evolution; in general, “homoplasy”).

And so we can ask: Is all change in the evolutionary histories of material cultural systems a matter of the transformation of preexisting states of homologous features? For clearly some of it must be: When the bell of cornets came to be lengthened (arguably in response to the market inroads of the rival trumpet—beginning in the early decades of the twentieth century), it was a progressive modification of a preexisting structure.

But consider the replacement of the original cornet piston valve by what has remained the “modern” valve. The original Stölzel valve, in which the airway ran vertically along the inside of the valve, was “challenged” and eventually replaced by the newer Périnet valve—which transmits the airway only across the valve. The first piston valves (ca. 1825) were all Stölzel valves; the oldest known Périnet-valved cornet was built by Adolphe Sax in 1843. It took until at least World War I for the Stölzel valve entirely to disappear—though it was largely replaced by the Périnet valve by the mid-1850s in all but the cheapest of instruments. The “transition” in the 1840s and 1850s even included some instruments with a mixture of both types of valves.

At issue here is the simple question: Did the Périnet valve “evolve” in some meaningful sense from the earlier Stölzel valve? Answer: No—not in the sense that the Périnet valve could be construed as a direct derivative, a descendant of the Stölzel valve. Rather, clearly it is an alternative design, meant to remove the sharp angles of the direction of the airway, as claimed in Périnet’s patent of 1838. Homologous, perhaps, as tubular piston valves—but the newer valve was Périnet’s way of thinking around what he saw as deficiencies in the structure—hence function—of the older Stölzel valve.

This is the “Hannah Principle” (Eldredge 2006; see Tëmkin and Eldredge 2007): In the “evolution” of material culture, new ideas, conceived as better ways to build an artifact (or at the very least to get around patent restrictions). Such considerations broach the topics of “directed variation” and indeed a sort of instantaneous selection. What is relevant here is that changes in material cultural systems that do not reflect simple modification of preexisting structures5 further complexify the problem of historical reconstruction—in this case especially through the routine application of the kinds of phylogenetic algorithms developed and used for biological systems. Once again, we see how evolution in material cultural systems is inherently more complex than it is in biological systems.

Hierarchies and Patterns

Phylogenetic reconstruction perforce focuses solely on the information-based system variably conserved and modified in the passage of time in both biological and material cultural systems. Folding in the economic context in which patterns of stasis and change occur allows analysis of the dynamics of such pattern generation through time—the heart and soul of true evolutionary theory.

As other authors (e.g., Rosenberg especially) have discussed in this volume, the hierarchical structure of social systems is manifest—as it is in biological systems, and, for that matter, material cultural systems. Analyzing the hierarchical structure of biological systems (e.g., Eldredge and Salthe 1984; see Eldredge 2008 for a concise review) has proved fruitful in developing theories of the biological evolutionary process on a variety of spatiotemporal scales—from “micro-” to “macro-evolution.”

The basis of the dual biological hierarchy scheme followed here rests on the ontological supposition that organisms engage in two—and only two—classes of processes: matter/energy transfer processes (“economics”)—that allow differentiation, growth, and maintenance of the soma (aka “staying alive”), on the one hand, and reproduction on the other. Both sets of processes generate hierarchical systems of which organisms are a part: Organisms are parts of local economic systems (“avatars”), which form parts of local ecosystems. Local ecosystems are connected geographically with adjacent ecosystems, across the boundaries of which matter and energy commonly flow. This is the “ecological hierarchy.” (Fig. 11.3).

Fig. 11.3
figure 12_3_185909_1_En

The ecological hierarchy

Full size image

Reproduction sets up a parallel hierarchy: Sexually-reproducing organisms are parts of demes, which themselves are parts of species. Species also “reproduce”—in the sense that speciation occurs, creating lineages of ancestral and descendant species that are themselves hierarchically arrayed as subsets within larger sets: the variously ranked taxa of the Linnaean hierarchy. This is the “evolutionary” hierarchy (Fig. 11.4).

Fig. 11.4
figure 12_4_185909_1_En

The evolutionary hierarchy

Full size image

Note how Darwin’s original formulation of natural selection virtually falls out of the mere act of setting the two hierarchies side by side. Organisms are of course the only entities in both systems: Natural selection is the statistical effect that relative economic success has on reproductive success among conspecific organisms (members of the same avatar/deme)—given the presence of heritable variation.

The “sloshing bucket” (Eldredge 2003a, 2008) notion of (macro-) evolution—i.e., at spatiotemporal scales larger/higher than within-population microevolution through natural selection and drift—is derived simply by mapping classes of evolutionary events onto the dual-hierarchy conceptual structure. Lower levels of ecological disturbance characteristically trigger recovery through ecological succession on the scale of months/years/decades. Recruitment comes from adjacent demes—and little or no discernible morphological evolutionary change accrues.

On the upper end of the spectrum, the five or six global mass extinctions that have occurred in the past .5 billion years are sufficiently devastating—accounting for the loss of so many species—that entire clades disappear. Recovery entails considerable evolution, as surviving clades often radiate (usually after a lag of several million years). The most famous example is the Paleocene and Eocene multiphasic radiation of mammals after the demise of terrestrial dinosaurs in the end-Cretaceous mass extinction event. Mammals had been present since the Triassic—evolving at roughly the same time as had the terrestrial dinosaurs (and other collateral reptilian kin clades). Yet, ecologically speaking, at least, mammals did not diversify to any major degree until after the dinosaurs had become extinct.

Thus radiations, involving extensive amounts of evolution occurring in bursts, are as a rule typically “decoupled” from the actual origin of major clades—marked as they often are by major adaptive “innovations.” Three middle ear bones, the presence of hair, lactation, and (later, but still in a pre-Tertiary world) the acquisition of placentation—mammalian hallmarks all—did not themselves trigger the bursts of evolution that produced most of the extant orders of mammals.

The spatiotemporally intermediate level of the “sloshing bucket” is in many ways the most interesting: When climate change or other environmental factors provoke extinction events of sufficient severity, covering sufficiently large spatial areas that the geographic ranges of entire species are affected, species begin to go extinct. Relatively rapidly occurring waves of extinction are common in the history of life, always followed by waves of speciation—mini-versions of the more notorious and easily appreciated mass extinction events. These are the “turnovers” of Vrba (1985). There is growing consensus among paleontologists that most speciation events are concentrated in pulses following regional extinction events (e.g. see van Dam 2006, on rodent evolution in Spain over the past 24.5 million years). And, if it is indeed the case that stasis characterizes most of the histories of most species in the history of life (Eldredge and Gould 1972), the implication is that most morphological evolutionary change occurs in conjunction with speciation events—a conclusion now gaining a measure of support from recent results in molecular biology (see Pagel, et al., 2006).

Most speciation events associated with turnovers produce only small-scale, relatively minor morphological change; rarely does the truly new appear in turnover-related speciation events. Yet, if the macroevolutionary larger scaled radiations following mass extinctions themselves usually fail to produce major-large-scale adaptive innovations that are justly considered to constitute the other half of what is usually meant by the phrase “macro-evolution”—it must be the case that many such “major” morphological changes in fact do occur in speciation events in the course of relatively modest “turnover pulse” events. Here, perhaps, is where relatively minor changes in the genetic regulatory apparatus, mediated by selection, produce the sorts of changes that, in hindsight at least, are recognized as the major large-scale changes of macroevolution.

Hierarchies in Material Cultural Systems

What would a hierarchy scheme of material cultural systems look like? Asking myself that question specifically as applied to cornets, I derived the hierarchical scheme of Fig. 11.5. At a glance, and once again, it appears that material cultural systems are at least an order of magnitude more complex than the much simpler dual hierarchy scheme that describes biological systems. Nor does the scheme of Fig. 11.5 display all the true complexities of the material cultural realm.

Fig. 11.5
figure 12_5_185909_1_En

Hierarchical structure of complex material cultural systems in an evolutionary context. For discussion, see text

Full size image

Like the biological systems, there are “pure” economic and informational hierarchies in material cultural systems: The regionally structured “marketplace” encapsulates the economic side—related to, but different from, the complexities of the world of music—where the instruments are of course put to use. On the other hand, like Plato’s chair, there is the “pure information”—the hierarchically structured world of artifacts, with their underlying information. Unsurprisingly, given the discussion above, these hierarchies are by no means as neat and unambiguous as are the hierarchies of relationships among sexually reproducing organisms. Cornets, for example, are members of the lip-blown “family” of musical instruments—often referred to as “brasswinds.” But unambiguous classifications of musical instruments are notoriously difficult to produce: Is a 4.5’ long brasswind equipped with a cornet-style mouthpiece, but with a slide instead of valves, to be called a “slide cornet” or a “soprano trombone?” Both terms have been used—depending to some extent on the use to which the instrument is put. It is a trombone in a trombone quartet; it was a slide cornet when Louis Armstrong used one occasionally with his jazz group in the 1920s.

Thus the information and economic hierarchies of biological systems do seem to have direct analogues in material cultural systems—though they are seemingly more complex in the latter than in the former. What is truly new about material cultural systems (vis à vis biological systems) is the dual hierarchy scheme lodged between the information and the marketplace: the manufacturers—themselves typically arrayed in local and regional hierarchies.

Minimally, of course, a manufacturing system requires a single maker—unlikely in the industrialized world of manufactured goods, but less unlikely in smaller, nonindustrialized social contexts—such as hunter-gatherer bands. No real hierarchy there, if only one person is involved; but there is still an interplay between information/ideation and the manufacture of a given object. Along with other lines of evidence, the work sites of stone tool traditions are identified as such primarily from the discarded shards produced in manufactory, as well as unused raw materials (which may have been rejected as not optimal—and, of course, as tools that did not turn out well enough to be used). Even if only one maker is involved, there is an interplay between the idea (i.e. the underlying information of the desired product and the steps known to be necessary in its successful production) and the actual production itself—which includes obtaining the raw materials and the actual labor involved.

In postindustrial manufacturing systems, many makers, employing several to hundreds of variously skilled people, are found in local and regional settings. My piston-valved cornet database lists some 125 makers (between the years 1825-Present), primarily located in France, Belgium, England, and the United States (northern and eastern European cornets have historically typically been fitted with rotary valves). Surviving instruments, advertising, patents, and written records help pinpoint the history of design innovation and the spread of information within and among manufacturers throughout cornet history (a far better record than anything I have seen in paleontology; fossils do not come with serial numbers!).

I have utilized 17 variables to describe cornet morphology; these variables are sufficient to describe all but a fraction of known cornet variation throughout the 180+ years of its “evolution.” As already noted, these 17 variables are almost completely “unlinked”: the “bell” (flaring portion emitting the sound) can be on either side of the valve cluster for example—and the configuration of the valve tubing can be fitted to any variant version of valves. As shown especially clearly in Fig. 11.1, the earlier Stölzel-valved cornets (often called “cornopeans”) were retrofitted with later innovations associated initially with later Périnet-valved cornets—including placing the bell to the left side of the valve cluster and moving the tubing from the second valve to the right side of the instrument; both conditions have long been invariant throughout later cornet history.

It is possible to see innovations arising in one atelier and being applied to other models under production in the same shop. The oldest known cornet with the modern configuration of the bell on the left-hand side of the valves was made by the Parisian maker Antoine Courtois in 1855. Called the “nouveau modèle,” shortly thereafter the same maker moved the bell to the left side on several different models—including one with a combination of Stölzel and Périnet valves, and on another two models that differed from the original configuration of the 1855 instrument. These latter two models rapidly became stabilized as the twin iconic models for the remainder of the nineteenth century. All four were called the “new model”—signifying the rapid spread of the idea within the shop itself.

The reason for this spread was the commercial success of these French-made cornets in London. England had used mostly German-designed (and/or made) Stölzel-valved cornets through the 1840s, but a switch to French-made instruments was already underway, imported by the Distin family, John Pask and perhaps others (As a youth, William Thomson, the future Lord Kelvin, who annoyed Darwin by insisting on a relatively young age of the earth, played a John Pask French-made Stölzel-valved cornet, possibly made by Antoine Courtois). Courtois’ “New Model” cornets (so stamped in English on London-sold instruments) were an almost instant hit in London. Played and endorsed by the virtuoso Hermann Koenig, star soloist (arguably history’s first instrumentalist “rock star”) in Jullien’s orchestra that mostly played in London, by the late 1850s cornets with the new “English bell” had all but replaced the older, traditional, retrospectively dubbed “modèle français”—“French model”—with its bell on the right side of the valve cluster.

Other makers began to catch on but not until nearly an entire decade had passed. Henry Distin had begun making his own instruments, and Courtois’ great Parisian rival Gustave Besson had opened a factory in 1858 in London: Both were making English bell instruments by the mid-late 1860s for sale to the English market. But this “new model” was much slower to become popular in France and other countries in continental Europe, not fully supplanting the modèle français until World War I.

An important feature of the hierarchy of makers that stands between, and connects, the hierarchies of “pure” information and economics (“marketplace”) is that it, in itself, is actually a dual structure of information and economics (Fig. 11.5). Part of an operation is concerned with innovation and production of existing models using plans (e.g. shop drawings and specifications, prototypes, patents, other makers’ models, and the experience of the artisans); the other part is concerned with procuring materials, equipping the factory, paying the workers, and advertising and marketing of the product in general. These two facets of course interact—as when decisions are made (based at least in part on sales results) on which models to produce, including whether or not to produce a relatively greater or narrower variety, and of differing quality and “price points.”

And, of course, information is transferred among makers—locally and regionally—driven in large measure by relative success of models in the marketplace. I have already mentioned the origin and spread of the English bell model of cornets—from regional hit in London to today’s worldwide design standard. Looking for all the world like the spread of an evolutionary novelty across a Sewall Wright-style landscape (e.g. Spencer, this volume), the theft of ideas in material culture (between makers, as well as between models in a single maker’s atelier) is an underlying dynamic of cornet evolution with ramifications that transcend the mere muddying of phylogenetic diagrams. Though there was a measurable lag in the spread of English bell designs, the mere fact that the information could be readily co-opted (and there were no known patents protecting the design in any country) accounts for a very different evolutionary history of the cornet than would have occurred had that design been restricted (like genetically based innovations) to a single small French company.

Figure 11.5 points to two distinct forms of “selection” in material cultural evolution. One is inherent in the very act of design novelty: A designer can conceive of a novelty and instantly reject it as impractical for a variety of reasons—without bothering to build a prototype, let alone testing it in the marketplace. The history of cornet design as seen in patents is very different from the actual, realized history—as makers routinely patented ideas that never got beyond the conceptual stage, or prototype stage, all the way to the marketplace. There is an element of selection to the very act of design creation—which otherwise is a directed form of variation, an ad hoc solution to a perceived problem, perhaps, that, if it seems plausible, might be further put to the test of prototype construction, then to limited production and marketing, etc. A form of “selection” enters in each step of the way (all subsumed under “Selection I” in Fig. 11.5). There would seem to be no analogue of “fitness” here—the “selection” merely being the evaluation of the designer(s).

The marketplace presents another, very different form of selection, where a form of “fitness” measured in sales, numbers of instruments—or even numbers of different makers producing the same model—might constitute a plausible analogue of the forms of fitness encountered in population genetics. Certainly continued existence—and expansion—of makers heavily depends upon their success in the marketplace. This—plus the unpredictable fluctuations of regional and global economies—are subsumed under “Selection II” in Fig. 11.5. But a caveat is necessary here: “fitness” in this sense has to do with the survival of business entities and not, in any formal, explicit way, with the relative reproductive success of owners, managers, and workers in those businesses.

Omitted from that diagram is another form of selection—as when makers decide to borrow ideas from one another. Makers compete with one another, but they also cooperate (some makers buy parts from others; price rigging is not unknown, etc.). What gets made, and what never makes it to the marketplace, is in part an inter-maker form of selection as well. When it comes to humanly made artifacts, selection is a multifaceted “process” that far transcends in its totality of complexity even the multiplicity of generally accepted forms of “selection” in biological evolutionary theory.

Material Cultural Evolution and the “Sloshing Bucket”

Is there an analogue to the “sloshing bucket,” where we can match classes of events in cornet design history to various spatiotemporal levels of this hierarchical structure of material culture? And, if so, are major evolutionary events initiated more by external perturbations in the economic arena, or are they a reflection of innovation and selection or a combination of factors from both the informational and economic sides of the hierarchical ledger?

Elsewhere I have documented the major features of cornet “evolution”—including examples of a number of evolutionary patterns, including gradual linear change, stasis, turnovers (of varying scales) etc. (Eldredge 1997, 2000, 2002). Here I will just briefly mention five major episodes in cornet history and see how they connect with the hierarchical scheme outlined above:

  • Valved Cornet Origins—ca. 1825. Brass wind instruments lacking valves or other means of changing the length of the tubing (e.g. slides, keys) can only play the overtone series: Limited by the principles of acoustical physics, it is impossible to play a chromatic scale except in the very highest reaches of the instrument attainable only by virtuosi. Made possible by advances in metallurgy, various devices (such as valves) that minimized leakage of air allowed inventors to try out a multiplicity of mechanisms that could change the length of the instrument—thus allowing full chromaticism (especially when a third valve was added a few years after the earliest two-valved cornets were marketed). This was an achievement driven by the needs of Western music that was enabled by general scientific and engineering advances. It was an instant success—though the valveless bugle maintains its role as a military signaling device.

  • Replacement of the Stölzel valve (invented ca. 1825) by the Périnet valve (invented ca. 1838). The alleged superiority (in production of sound qualities) of the Périnet valve over the Stölzel valve was not immediately apparent to musicians. Manufacturers continued to produce Stölzel-valved instruments alongside Périnet-valved instruments (as well as a “hybrid” form with a central Périnet valve flanked by two Stölzel valves) well into the 1850s. The “hybrid” form allowed the classic deep-bodied form of the older Stölzel-valve cornopeans to be retained. Tests with restored instruments reveal no discernible superiority in sound qualities or playing characteristics. The best explanation of the eventual replacement of Stölzel-valved instruments by Périnet instruments is that makers evidently found them to be a sturdier form of construction and less prone to wear and tear (Robb Stewart, pers. comm.). Moreover, there has been no experimental or experiential confirmation of the near-constantly made claims of rival manufacturers that they had devised a type of Périnet valve superior to all others in acoustical properties. The public, choosing from the variety of styles, including valve types, seems to have made its choice on reputation of maker and overall style (i.e., appearance) of an instrument.

  • Such may be the explanation for the instant success of the “English bell” in England: the novel look of the instrument. However, with the bell on the left side of the valves, there is a more secure purchase as the player grips the valves with the left hand.

  • 1900–1905: Major turnover in cornet design. The iconic Victorian double waterkey designs of Courtois (copied by many other makers) almost completely disappeared, virtually overnight—replaced by a slight modification of a more conservative design dating from the late 1860s/early 1870s. Nothing about the replacement models were alleged to be superior to the Victorian models they so abruptly replaced in the public’s esteem. Nor were there global depressions, changes in music or anything else of a large-scale nature that could be reasonably supposed to have caused this turnover, except one thing—the change in the calendar from 1899 to 1900. Out with the old—Victorian style—in with the new! Advertising of all manner of manufactured goods reflected the need for novelty for the new century. In the United States (unsurprisingly), Conn, York, Holton, White (King), Buescher, and other makers outdid themselves producing many newfangled, short-lived models—most pronounced to be superior in their playing aspects (these included still more experimental valve designs). Planned obsolescence was part of this, as many models (especially by Conn) were continuously “improved” with each passing year. What won out—the dominant cornet model still in use today—was however a simple style invented by Besson in the late 1860s—with some later, minor modifications.

  • Near extinction of the cornet—1920s. Cornets, especially in the United States, had become gradually longer, more trumpet-like, throughout the first two decades of the twentieth century. Sometime in the early to mid-1920s, the modern form of the Bb trumpet—invented by cornet makers (probably the aforementioned Parisian maker Besson) in the late 1870s/1880s through an experimental trial-and-error phase—suddenly became popular not only in jazz but in commercial and symphony orchestras and town bands. Bb trumpets had never been sold by Sears, Roebuck until the early 1920s. Why the chief symbol of this switch—Louis Armstrong—himself dropped the cornet for the trumpet in the 1920s is not really known: Some authors feel that the exigencies of recorded music might have had something to do with it. Trumpets are brighter instruments, capable of producing louder, more piercing sounds than the traditional more-horn-like, mellower cornets—with their deeper mouthpieces. But I agree with Lewis (1991) who has pointed to the “trumpetization” of cornets for several decades prior to the switch—a reflection (he plausibly argues) of the preference for lighter, brighter upper register sounds from the brass section that can be traced back at least as far as the bands and music of the American Civil War.

Thus the major features of cornet evolutionary history seem to overwhelmingly reflect marketplace fickleness—an unanticipated (i.e., by the makers) rejection of what had been “the usual” in favor of something else already available out there. The exceptions are the relatively few true “improvements” in design history: addition of valves for chromaticism; less dramatically (and consequently more slowly) adoption of the “English bell.” But the near-extinction of Victorian double-waterkey cornets at the turn of the past century; and the switch to Bb trumpets that nearly drove cornets completely extinct (especially when trumpets are considered a derivative clade off cornets in the first place) reveal a pattern of external perturbation not driven by functional considerations of musical worth, or indeed by changes in musical style, so much as by shifting moods in the marketplace. The end-century switch-over seems to have reflected nothing more (nor less) than a desire to have something more modern—even though what was chosen was basically a tinkered version of an alternative model that had already existed for three decades. And the sudden adoption of the Bb trumpet was a culmination of a trend in that direction—the “trumpetization of the cornet”—finally just going the final step, and adopting something that had been available, but little used, for half a century. If it reflected musical taste, it was the taste of the musicians and the public they played for and not of the composers of any new form of music that called for brighter sounds over the mellower cornet sounds of old.

Conclusions

Like biological macroevolution, then, material cultural evolution seems seldom driven by “key innovations”—the better mousetraps that the consumer world is famously supposed to be beating a path to. It seems more driven by stylish fads—which sometimes reflect a true design change (as when clamshell flip phones surprised industry giant Nokia, losing them a major market share). Yet even here, nobody thought the clamshell phones worked better as phones than the older, more traditionally telephone-like cell phone models: That, after all, depends on the electronic guts of the phone—and on the carrier which it connects to. But the flip phone is cooler.

As long as a tool works, it stays in the ambient culture. True innovations that actually improve tools are very rare—some of the best cornets I have ever played were made in the 1850s. If the look, the feel—the gestalt, the jizz—of a tool can be changed without compromising its function, that will happen as unanticipated rapid turnovers.

And, of course, it is at the very beginning of the existence of classes of material cultural objects that the interplay between the creative mind of the true innovator and the successful adoption and spread of the innovation is most clearly seen. It is not the better mousetrap, but the initial fact of a mousetrap. And even the invention of the truly new—like computers or even desktop computers—may take a while to catch on.

These isolated events taken from cornet history are merely suggestive. Do they constitute actual patterns—classes of events that permeate the history—not just of cornets, but of other musical instruments, and for that matter, all the evolution of all material culture? For focusing on the evolutionary history of one well-delineated system over its entire history, informative as it may be, would be like deriving an entire theory from a single example. The “sloshing bucket” approach to macroevolutionary history could not be derived in full from the history of the Class Trilobita: Patterns of disturbance-induced extinction and subsequent evolutionary diversification within surviving lineages need to be timeless (e.g. seen throughout at least the Phanerozoic history of multicellular life) and affecting at least potentially all forms of life—not just a single clade.

Even granted the complex hierarchical system underlying the manufacture and sale of cornets to be potentially general for industrially manufactured goods, the critical question is: Were there similar events affecting other material cultural products at the same time and place? My claim, for example, that the turn of the century in itself led to the shift away from the ornate Victorian models to more conservative, yet more streamlined, models predicts that cornets are but one example of a general fin-de-siècle/brave-new-world mentality in the marketing world/cultural life of Western Europeans and Americans generally. There literally have to be more examples.

And what about the near extinction of cornets as the Western world (save the British Brass bands) nearly universally dropped the cornet in favor of the Bb trumpet (which itself hardly underwent any further “evolution”—it was ready to go right off the design shelf). This sounds like it could be a unique event, but what else happened in the Western cultural world of musical instruments in the 1920s? Vague allusions to unspecified acoustical demands of the recording (and radio) industries, unsatisfying as they are, would imply that perhaps other classes of instruments were similarly affected. There was the famous “saxophone craze” in the 1920s: There are as many old saxophones in attics in the United States as there are violins—and cornets. Maybe more. And the possibility arises that this sudden mania for saxophones (which had been around—virtually unchanged, but little used—since Adolphe Sax invented them in the 1840s) and the sudden rise of the Bb trumpet had the same underlying cause: if not the demands of the electronic era, then perhaps the marketing genius of Carl Greenleaf, who took over the floundering Conn corporation in 1916 and, faced with the radio-induced demise of the local town bands (there had been over 80,000 such bands in the late nineteenth century according to Hazen and Hazen, 1987), invented the “school band movement” that saved his company—and gave a shot in the arm to the entire industry. Perhaps the saxophone craze and sudden demise of the cornet in favor of the Bb trumpet represent successful outcomes of a deliberate marketing scheme.

Evolutionary patterns—like stasis, gradual change, and turnovers at various spatiotemporal scales—do seem to be common to both biological and material cultural evolutionary realms (see also Prentiss this volume; Kuijt and Prentiss this volume). The greater complexity of material cultural systems derives in large measure from the greater frequency of horizontal information transfer in material cultural systems, and from the violation of the principle of homology engendered by the “directed variation” of ad hoc solutions dreamt up by a fertile creative designing mind—the “Hannah Principle.”

And both sorts of systems are manifestly hierarchically arranged—but the presence of those creative minds establishes a dual information/economic hierarchy in between the “pure” hierarchies of information and economics in human material cultural systems. The “sloshing bucket” notion of biological evolution—which relates disturbances and extinctions at varying spatiotemporal scales to magnitude of evolutionary events—may have its analogues in the evolution of material cultural systems.

The “sloshing bucket” further suggests that evolutionary rebounds after particularly large-scale extinction events are not necessarily where the more profound adaptive innovations in biological evolutionary history necessarily arise—further implying that those novelties in themselves seldom act immediately as “key innovations” triggering prodigious amounts of rapid diversification. The same appears to be true in material cultural systems (Chatters this volume; Zeder this volume): Most innovation in cornet design history come early in the histories of the ateliers where they first arose. If “English bell” cornets were an instant hit in England (though not in France where they arose in Courtois’ shop in the mid-1850s), the success of the Besson desideratum (forerunner to the modern cornet, invented in the 1860s but not market-dominant until the twentieth century) and the piston-valved Bb trumpet (derived from the cornet in the 1870s but not market-dominant until the 1920s) are the more typical cases in point.

But all this remains to be seen as material cultural systems over the last 2.5 million years are looked at with these possibilities in mind.

Notes

  1. 1.

    In this paper, I shall be drawing on my database on the history of piston-valved cornets—lip-blown brasswind instruments manufactured and sold between 1825 and the Present, predominantly in France, England and (somewhat later) the United States. The database consists of 17 variable vectors that describe cornet morphology and comprises nearly 200 makers and over 120 distinct “models” (i.e. unique vectorial combinations).

  2. 2.

    not excepting humans—though culture has radically altered the straightforward usual correlation between economic and reproductive success, with exceptions that invariably delight sociobiology’s proponents, robber barons, and the just plain wealthy typically have fewer children than do people in lower socioeconomic classes—a point well developed, of course, by Karl Marx.

  3. 3.

    Though long ago George Simpson denied that differences in generation time neatly transcribes into differences in the rate of evolution, when he maintained that elephants, with their long generation times, appear to have evolved more morphological disparity than have the far shorter-generation mice—over the same lengths of geological time.

  4. 4.

    Molecular biology has recently—and very interestingly—muddied the waters on the very concept of homology. It turns out the same genetic basis for the vertebrate eye—the Hox gene—is also what is responsible for the development of arthropod eyes. Traditionally, vertebrate and arthropod eyes have been considered excellent examples of analogous (nonhomologous) structures—what with their manifestly separate phylogenetic development and utterly different conformations. Yet now we hear they are, at some even deeper genetic level, homologous! Fun!

  5. 5.

    Mme. Besson, highly successful heir of her late husband Gustave’s musical instrument-making atelier, patented yet another form of the Périnet valve in Paris in 1874. The patent text is fascinating—as Mme. Besson gives a (largely accurate) history of the Périnet valve from its inception through her present patent design (Eldredge 2003b). Tellingly, though she traces the development (I would say “evolution”) of the valve through four antecedent stages, she pointedly does not refer to the Stölzel valve—clearly to her an utterly different valve form altogether.