To recap, the gist of that post is that the Indo-European hypothesis is large and contains multitudes and that the options seem to be to accept the astonishing inexactness of the metaphors or submit to the paralyzing mind-blowingness of what we use them to try to explain. I suggested that the latter option could be inconvenient if you're trying to discuss historical linguistics and language relatedness in a class that meets for an hour and fifty minutes twice a week for 15 weeks.
Anyway, continuing on the topic of the metaphors that we use to try to create some kind of manageable order out of the chaos that is the story of human language and how it got this way, we turn now to a fellow name of August Schleicher (1821-1868), a German linguist by training and profession who specialized in classical and Slavic languages. Schleicher, who may have had some of the same concerns that I have about how we can possibly even try to conceptualize an unattested 5,000 to 7,000-year-old super-ancestor Ur-language that might not even have actually existed, decided that it was time someone got around to the task of trying to reconstruct Proto-Indo-European.
That meant recreating (creating?) more or less an entire language -- vocabulary, phonology, grammar -- by working backwards from existing linguistic data found in the oldest surviving texts in languages believed to be descended from PIE.
Despite the seeming complete and utter impossibility of such a task, Schleicher actually did it. I can't believe no one tried to talk him out of it ("Gus, dude, that is völlig bekloppt!"), or if anyone did try, that he resisted and did it anyway. He actually reconstructed Proto-Indo-European, a language that left no direct evidence, if it had ever even really existed, and if it had existed, it had been dead for something to the tune of 5,000 years. Let that sink in for a minute.
And if you're not blown away at the thought of the kind of brain Herr Doktor Schleicher must have had to pull this off, go back and read my previous post, especially the parts about using the comparative method for reconstructing languages with no living speakers and no direct textual evidence. It's important to me that everyone understand that this was a feat of extraordinary intellectual bad-assery. (That it was also a feat of extraordinary nuttiness is not necessarily beside the point.)
Anyway, in 1861, Schleicher published his reconstruction of PIE in a book called Compendium der Vergleichenden Grammatik der Indogermanischen Sprachen, known in English (and available in translation here) as A Compendium of the Comparative Grammar of the Indo-European Languages. Revisions and reissues appeared well into the 1870s, although Schleicher himself died in 1868 at age 47.
"What does all this have to do with metaphors?" you might be thinking. Everything. It has everything to do with metaphors. For one thing, even as Schleicher published his reconstruction of a 5- to 7,000-year-old dead language that might not have existed in the first place, he also made it clear that he knew all along that he was dealing in metaphors, and particularly in a big PIE-shaped metaphor, one that made it possible for him to reconstruct what was quite possibly a mythical language. As he wrote in the Compendium in 1861 (although of course he actually wrote it in German):
"A form traced back to the sound-grade of the Indo-European original language, we call a fundamental form. When we bring forward these fundamental forms, we do not assert that they really were once in existence." (Emphasis added.)I mean, duh.
But that's not all. As if the actual reconstruction of a possibly metaphorical language is not enough to guarantee Schleicher's place in history, or at least his place in historical linguistics, or at least in the history of metaphors to explain historical linguistics, there is also this: August Schleicher generated some of the most influential and enduring metaphors to which we have recourse today for making sense of the development of human language over time, including the single most influential and enduring metaphor of all: the phylogenetic tree for mapping language descent and relatedness.
That his tree metaphor has been criticized and challenged from practically day one, including by Schleicher protege (and total ingrate) Johannes Schmidt , and continues to be qualified to within inches of its life even today, ought to take nothing away from the fact that it is actually still used today. Schleicher devised his family tree theory to explain relationships among languages and thereby to classify them, although he actually called it Stammbaumtheorie, which is German for 'family tree theory' (sort of), because he was, you know, German.
Anyway, according to Richards (2002: 34), Schleicher "suggested (but did not yet graphically illustrate) that the developmental history of the European languages could best be portrayed in a Stammbaum, a stem-tree or developmental tree” as early as 1850. Richards notes that his first “graphic representation of a Stammbaum" appeared in two publications in 1853.
Unfortunately, neither of the two 1853 articles in which Schleicher's proto-tree proto-drawings (see what I did there?) first appeared is readily available, so I had no choice but to copy it, as in reproduce it by trying to draw it myself. While the image is conveniently reprinted in Richards's article, the article itself is actually Printed in a Book that is Protected by Copyright. Even though Schleicher's original work is of course in the public domain, reprinting it in a book in 2002 might give a publisher a sense that they are entitled to righteous indignation as well as possible legal recourse were someone to, say, scan the image and put it on the internet. Hence my original interpretation:
By 1860, Schleicher “had begun to use Stämmbaume rather frequently to illustrate language descent,” according to Richards (34). And his designs get more sophisticated as well in the 1860s, meaning that he seems to have used a ruler this time. You can see that in the illustrations below, which were included in the original German edition and the 1874 English translation, respectively, of the extremely compendious Compendium, which is long out of copyright and from which I can therefore freely reproduce the illustrations here. First, here it is in the original German version (1861):
First is the regularity hypothesis, which posits that speech sounds change over time in systematic (that is regular, predictable) ways, as the Danish linguist and astonishing brainiac Rasmus Rask (1787-1832) had earlier theorized. (There's more about Rask in my previous post.)
The second hypothesis is the relatedness hypothesis: Because of this (assumed) systematicity of change, sound similarities among particular languages are therefore likely to be the products (and evidence) of family relationships (sometimes called genetic relationships, to use another biological metaphor), among those languages.
All of this was pretty innovative thinking, and the best part is that had Schleicher not had an actual life outside his work at the university, he might never have come up with any of it.
In addition to his completely understandable passion for linguistics, Schleicher was also an enthusiastic gardener and avid reader of scientific literature on the topic. In an 1863 essay, “Darwinian Theory and the Science of Language" (which is essentially an open letter to his close friend and colleague at the University of Jena, the zoologist Ernst Haeckel), Schleicher outlines his thoughts in response to Darwin's On the Origin of Species (1859), which Haeckel had encouraged him to read when it was translated into German in 1860. Schleicher notes in the essay that Haeckel had recommended Darwin's book to him because he thought it would appeal to his linguist friend's love of gardening (which is translated adorably in the English translation of the essay as “botanizing"), but he writes that upon reading it, the British naturalist's “views and theory struck me in a much higher degree, when I applied them to the science of language."
As Schleicher explains in the "Darwinian Theory" essay, by the point he had reached in his career at the time of its writing, he had come to conceptualize languages in the context of the natural sciences, about which he read avidly and which developments he had followed with great interest for most of his life. In the essay, he maintains that human languages are essentially living organisms that are born, grow into maturity, and eventually die. This is a reasonable enough metaphor, even a pretty good one, but interestingly, and despite the apparent predisposition for metaphor that his reconstruction of PIE and invention of the Stammbaumtheorie might suggest, Schleicher does not appear to treat the language-as-organism metaphor as, you know, metaphorical.
He had once suggested as much, in a book he published in 1853, Die Sprachen Europas in systematischer Uebersicht (The Languages of Europe in Systematic Perspective -- no online English translation of this book available, unfortunately), but by 1863, he was no longer saying merely that languages are like living organisms but rather suggesting that languages actually are living organisms that share biological characteristics with plants and animals, at least in the evolutionary sense.
Lest readers misunderstand what he is saying here, he assures them that he does not mean to limit his claim for the biological, evolutionary nature of language to what was then becoming a fairly uncontroversial application of language development in the context of human physiological evolution (i.e., the role that evolution plays in the development of the physical apparatus that human beings use to produce speech). He takes a much more radical position than that, one that classifies language itself as living organism and proposes classifying all of human life according to its linguistic systems:
"I do not here exclusively refer to a physiological investigation of the various sounds of speech, a study which has made considerable progress of late, but also to the observation and application of linguistic varieties in their significance for the natural history of man. What if those linguistic varieties were to form the basis of a natural system concerning the unique genus homo? Is not the history of the formation and progress of speech the main aspect of that of the development of mankind? This much is certain, that a knowledge of linguistic relationship is absolutely requisite for anybody who wishes to obtain sound notions about the nature and being of man."A few pages later, he takes the claim of language-as-organism even further:
Darwinian Theory, p. 15.
"Languages are organisms of nature; they have never been directed by the will of man; they rose, and developed themselves according to definite laws; they grew old, and died out. They, too, are subject to that series of phenomena which we embrace under the name of “life.” The science of language is consequently a natural science; its method is generally altogether the same as that of any other natural science."Richards (2002: 47) maintains that Schleicher and Darwin, who corresponded, were mutually influential and that Schleicher's tree designs impressed the naturalist, who cited Schleicher in The Descent of Man (1871: 56) as a source for his exposition on the "origin of articulate language."
Darwinian Theory, pp. 20-21.
However, it turns out that Schleicher's position, that "The rules now, which Darwin lays down with regard to the species of animals and plants, are equally applicable to the organisms of languages" (Schleicher 1863: 30), which he seems to have meant literally, did not prove to be very persuasive to other historical linguists, although it has enjoyed some considerable success as -- you guessed it -- a metaphor.
So here we still are, still talking about the Indo-European hypothesis, and still using Schleicher's models of language relatedness and descendancy, still applying the language-as-organism metaphor and the family-tree model as ways of conceptualizing the otherwise unimaginable. Like Indo-European languages themselves, Schleicher's metaphors, the ones he intended as metaphors as well as the ones that just turned out to work better that way, have ended up having pretty serious staying power. And despite its limitations (it is not well suited to account for internal variation, gradual change over time, or contact between the branches that represent families or sub-families, for example), his central metaphorical symbol -- the tree -- has thrived beyond what the dedicated botanizer-linguist might ever have hoped or imagined. For evidence of that success, click here to see a selection of the practically infinite visual representations of the Indo-European language family, all of which are in the debt of one August Schleicher.
 Schmidt explains language relatedness according to a wave model (wellentheorie), which posits that linguistic innovations occur first at a single point and then diffuse outward over time in gradually weakening concentric circles. I like this model very much for thinking about language variation and change over time, but it has not displaced Schleicher's tree model for relatedness, which it really does not conceptualize very well.