When I use an evidence-based word . . .

March 30, 2016

Jeffrey Aronson, Honorary Consultant Physician & Clinical Pharmacologist

Jeffrey Aronson

2. The roots of language

Which words came first?

In his Historiai, Book II, Herodotus tells how an Egyptian king, Psamtik (he calls him Psammetichus), undertook an experiment. He entrusted two children to a herdsman, charging him to allow no one to utter a word in their presence, to keep them in a cottage, and to care for them, from time to time allowing goats into their apartment to supply them with milk. Two years passed, and one day, when the herdsman entered their room, the children ran to him and clearly said βηκος (bēkos), the Phrygian word for bread.

So was “bread” the first word? Well, Herodotus’s account of Psamtik’s linguistic experiment, true or not, makes good reading. But it seems more likely that their caprine contacts caused the children to say “beeeeeh-kos”? It may not be entirely coincidental that in Greek βηκία (bēkia), a word that was used by Hippocrates (as cited by Galen), meant little sheep. Baaaaah!

Speech is prehistoric, having emerged many thousands of years ago, and we cannot possibly know what the first spoken word was. There is even debate about who might have uttered it. Did it emerge with homo sapiens? Neanderthals were supposedly incapable of it, since they supposedly lacked the requisite anatomical apparatus, but the discovery that they possessed a hyoid bone suggests otherwise.

On the other hand, there is abundant linguistic and archaeological evidence of proto-languages, the precursors of those we speak today. Latin is an obvious direct forebear of modern Italian, and ancient Greek of modern Greek. And languages can develop in different ways from common ancestors—compare, for example, modern French and Canadian French or Portuguese and Brazilian. But proto-languages go back much further than any of these, being the earliest attested or hypothetically reconstructed forms of languages, including such gems as proto-Algonquian, proto-Athapaskan-Eyak, and proto-Hattic. Best known is proto-Indo-European, and its discovery contained a surprise: that there was a single linguistic forerunner of most Indian and European languages.

Sir William Jones, an English judge (pictured), whom Dr Johnson described as “the most enlightened of the sons of men”, came to Calcutta in 1786 and tackled Sanskrit—a tough task. As Monier Williams tells us on page 1 of his Sanskrit Grammar (2e, OUP, 1857), devanagari, the Sanskrit script, had 14 vowels (all but one having two forms), 33 simple consonants, and 400–500 conjunct consonants. But Jones mastered it and in doing so made his discovery: “No philologer could examine [Sanskrit, Greek, and Latin] without believing them to have sprung from some common source, which, perhaps, no longer exists.” He was right.

Sir William Jones (1746–94)

Sir William Jones (1746–94)


Thomas Young, discussing Johann Christoph Adelung’s Mithridates, oder Allgemeine Sprachenkunde (Quarterly Review, October 1813), called that common linguistic source “Indo-European”. And in 1916 Carl Darling Buck introduced the term “proto-Indo-European” to describe the theoretical roots of the language from which modern words are thought to have developed.

In future blog entries I shall often start the process of investigating a word by examining its Indo-European roots.

Let’s do this to see where the word “first” comes from. First, consider the basic hypothetical Indo-European root PER, meaning “through” or “in front”, the latter in two senses, front-running and confronting. This difference can be appreciated by comparing “persuasion” and “perdition”. PER had a range of different derivative forms. In Greek those forms variously yielded πρóς, in front, πᾰρά, alongside, and περί, around, and in Latin pro, before or for, praeter, beyond, and per, through. One’s profile features one’s face and prosopagnosia is an inability to recognize faces. Prose is straightforward writing, unlike verse (Latin vertere, to turn), which turns around line by line. William Prout hypothesized that hydrogen’s primordial positively charged particle was the basic matter of which all elements were composed; Rutherford called the particle a proton. Prolactin is secreted before lactation, progeria is precocious senility, a prodrug is a precursor of an active one, and preproparathyroid hormone has a lot of previous.

But that’s not all. Consonants change: plosive p in Greek and Latin becomes fricative v and f in Germanic. So PER also gives us forbid, foremost, before, former.

Now we are ready to investigate “first”. The cardinal forms of number words are “one”, “two”, three”, and so on; the ordinal forms are “first”, “second”, “third” … The cardinal forms of the word “one” in Indo-European languages generally derive from the root OINO, but its ordinal forms come from PER. In French the cardinal “un” has the ordinal counterpart “premier”; in Italian “uno” matches “primo”; in German it’s “ein” and “erste”. So in English the cardinal form is “one” but the ordinal is “first”. At last!

A previous version of this blog was published at http://blogs.bmj.com/bmj/2015/02/09/jeffrey-aronson-when-i-use-a-word-first-things-first.


Cite as Aronson J (2016): The Roots of Language. CEBMJ https://www.cebm.net/use-evidence-based-word/ DOI: 10.13140/RG.2.1.4171.5603 

Leave a Reply

Your e-mail address will not be published. Required fields are marked *

* Checkbox GDPR is required


I agree