I’ll assume that linguists for a very long time were aware of similarities between Latin and Greek. What steps along the way (that is, how long ago) ended up also linking them with Sanskrit, Persian, Slavic, Germanic, etc. as one big happy family? I’m not really looking at the methods of linguistic theory, or whatever it’s called, just a timeline of when people said, “Aha! Here’s another relative!” Did the Grimms contribute much in their collecting of fairy tales?
In the first century B.C., Romans had noticed the numerous similarities between Latin and Greek. These similarities were attributed to the Greek origin of the Latin-speaking people (cf. the Aeneid). So the Romans, in pre-Christian times, both recognized similarities between their language and another, and gave a genetic/historical explanation for this. Even though the explanation wasn’t quite right, it was a first step on the way.
In the Middle Ages, the Bible-based belief in the Tower of Babel meant that etymologists tried to derive all language from Hebrew. They had to take such phonetic and semantic liberties to force etymologies out of Hebrew for unrelated languages, it brought etymology into disrepute and held it back until the 19th century.
In 12th-century Iceland an anomynous text called the First Grammatical Treatise showed how Icelandic and English are related. Toward the end of the 12th century, Giraldus Cambrensis wrote that Welsh, Cornish, and Breton were descendants of a single earlier language, which he called “antiquum linguae Britannicae idioma.” Not only that, but Cambrensis went on to suggest a relationship between this “ancient Britannic tongue” and Latin and Greek, because of lexical resemblances such as Latin sal, Greek hals, and Welsh halen, ‘salt’.
Apparently Dante never realized that the similarities he saw in the Romance languages were the result of their all being descended from Latin. Although it’s hard for me to imagine how a scholar of Dante’s caliber could have missed the fact. Even though the principle of genetic relationship of languages hadn’t been established yet, Dante must have been able to see how Italian had developed out of Latin. When he wrote De vulgari eloquentia he compared Latin and Italian extensively.
Apparently the genetic hypothesis for the Romance languages was proposed in the first half of the 15th century. Poggio Bracciolini (1380-1459) and Leonardo Bruni (1370-1444) saw that the modern Romance tongues, including Romanian, were the result of linguistic change as it had affected Latin. Even so, this early recognition of a genetic hypothesis did not catch on at the time for other language groups.
Wordlists helped to show languages in relation to each other. Lots of wordlists in different languages were compiled, beginning with Mithridates in 1555. In the late 18th century, Catherine the Great of Russia commissioned P.S. Pallas to collect and organize wordlists from a large sample of the world’s languages. It was published from 1786 to 1789 and contained information on 200 languages. Probably it was Catherine herself who decided how to group the languages, and this was done largely on the basis of genetic relationships. She put Slavic languages first, then Celtic, Greek, Romance, Germanic, Baltic, Caucasian, Finno-Ugric, Iranian, Semitic, Turkic, and so on.
Coincidentally, this was first published the same year that Sir William Jones gave his famous talk to the Asiatick Society of Calcutta, the birth of Indo-European.
J.J. Scaliger had attempted to classify languages into genetic groups in 1610, based on the similarities in the word for ‘God’. He got all the groups right, but he missed the larger genetic relationship of the groups that belong to Indo-European. What Scaliger did that was important was mass comparison of many languages at once. People focusing too narrowly on one binary comparison at a time can miss the forest for the trees. In 1615 a Lithuanian named Micalon claimed a relationship between Lithuanian and Latin, though he missed the relationship between the more closely-related Lithuanian and Russian.
The first stirring of what Sir William Jones would eventually realize was in 1686: the Swedish scholar A. Jager gave a public lecture in which he pictured an ancient language spreading over Europe, breaking up into daughter languages, which in turn led to Persian, Greek, Romance, Slavic, Celtic, and Gothic.
The basic structure of the Uralic family had been worked out by von Strahlenberg in 1730 and von Schlözer in 1770. Also in 1770, the Hungarian priest J. Sajnovics connected Hungarian with Lapp, so the Uralic family was known before Jones discovered Indo-European. In 1778 William Marsden pointed out the relationship of the Polynesian languages to one another, and three years later added Malagasy. On this point, Marsden had been preceded by Hadrian Reland, who made a similar observation in 1706, and Frederick de Houtman, who had perceived the affinity of Malay and Malagasy as early as 1603.
In 1767 P. Courdoux asked the Académie Française for an explanation of similarities between Sanskrit, Greek, and Latin, similarities which he believed to be relics of mankind’s original language.
That Sanskrit displayed similarities with European languages had been known for 200 years before Jones, ever since the Italian Filippo Sassetti, who lived on the west coast of India in the 1580s, had reported striking resemblances between Sanskrit and Italian.
Finally in 1786 Jones attributed the similarities to a common origin (though he did not imply they went all the way back to the Tower of Babel). All the pieces of the puzzle were there before Jones came up with his breakthrough synthesis. What Jones did that earlier scholars had missed was to collect, organize, and evaluate all of the evidence at once instead of limiting oneself to binary comparisons. Jones’s hypothesis eventually led the way for serious etymological research. It took until 1814 when the Danish scholar Rasmus Rask went to work on the topic and began to draw up the regular sound correspondences between the various members of Indo-European. Rask based his work on Germanic, Latin, Greek, Slavic, and Baltic (but unfortunately he left out Sanskrit and Celtic; he changed his mind a few years later and added Sanskrit and Persian). In 1816 the German Franz Bopp was able to correlate the abundance of similarites in the inflectional endings. Also, importantly, he showed that Sanskrit was not the origin of the branches of Indo-European, but was itself one of the branches. Jacob Grimm’s 1819 grammar established systematic sound correspondences, although the idea had been originated by Rask, including the sound changes that Max Müller named “Grimm’s Law.” In 1861 August Schleicher was the first to try reconstructing proto-Indo-European, and he also was the first to accurately work out the inner taxonomy (the lower-level groupings of how the branches fit together).
I think Rask gets the credit for having coined the name “Indo-European” in 1814.
Johanna, thanks for this very thorough and impressive answer.
My sentiments exactly. I knew about 20% of what Johanna said, but with no overall coherency to my understanding. That was a brilliantly thorough and concise summary, making a complex matter clear. Thanks!
To point out in passing, referencing Jakob and Wilhelm Grimm: They were not hobbyists collecting folk stories, but rather what would today be called linguistic anthropologists, studying the origins of Germanic language and culture, and perhaps among the foremost scholars of that erudite discipline anywhere and at any time. Grimm’s Fairy Tales, for which they’re best known to the general public, was their collection of old Germanic folk tales in aid of this scholarship.
Thanks, folks. 
I cribbed the above information from a book by Merritt Ruhlen, A Guide to the World’s Languages. Volume 1: Classification, published by Stanford University Press. This book is valuable for narrating the story of how linguists recognized and figured out the taxonomies for all the language families of the world. It contains a historiography of linguistic genetic taxonomy, just the very thing the OP was asking about. It also gives family trees for about 5,000 languages.
The 19th-century Romantic era was when searching for the roots of one’s folk culture came into vogue. It was of a piece with the discovery of ethnic nationalism (Greek War of Independence against Turkey, Kossuth’s Hungarian nationalist rebellion against the Hapsburgs, Lönnrot’s collecting the Kalevala in the backwoods farmsteads of Karelia, Smetana’s Bohemian rhapsodies like Ma Vlast, etc.).
My OED (ist ed.) gives the credit for naming Indo-European to T. Young in the Quarterly Review in 1814. It’s possible that they wanted to give prefernce to a Brit rather than a Dane, but that’s just a WAG.
IIRC (and IRvaguely here), Rask used “Indo-Germanic” as his phyletic term, and it was soon (presumably by the T. Young you cite) changed to “Indo-European,” which has been consistently used (except for a few people using “Indo-Hittite” for a while, an altogether odd story) ever since.
Thanks for the correction. That was one bit of information that Ruhlen left out: who first came up with the name Indo-European. I found from another source that the term dates from 1814, which is why I guessed it might have been Rask, although I seemed to remember that Indo-Germanic was Rask’s coinage. Ruhlen makes no mention of T. Young. The Continental Germanic speakers would have to come up with Indo-Germanic, wouldn’t they? I always thought it was the Brits who chose the more inclusive term.
Calling it “Indo-Hittite” has to do with a question of where to put Hittite taxonomically.
Ruhlen still uses the term “Indo-Hittite” although he admits “most linguists today reject it outright … Anatolian is for most scholars simply a coordinate branch of I-E on a par with Italic, Celtic, Germanic, Albanian, etc.”
My sentiments exactly. Thanks a bunch!
To throw out the traditional joke when we discuss this stuff, Schleicher apocryphally did this during a sabbatical year. As he left, his colleagues wished him well with the valediction:
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
“When come back, bring P.I.E.”
::: ducks and runs ::::