The word hybrid (from Latin hybrida, ‘mongrel’) commonly refers to animals and plants of mixed lineage, and more recently to vehicles with two or more power sources. In linguistic morphology it refers to a word formed by combining elements that originated in two or more languages. The process is called hybridization.
Many new words arise through compounding and affixation, and a lot of roots and affixes in English derive from Latin or Greek — sometimes indirectly, such as through French. (Classical compounds are a related source of new vocabulary, but they are of a ‘purer’ strain than hybrids and need not concern us here.)
There is a tendency for like to join with like, but because affixes from other languages are so well-established in English, and their origins are not widely known, etymological affinity is not routinely observed when words are formed. English has always added foreign bits to native bits, and both to other foreign bits. It does this in its sleep.
Hybrids are ubiquitous: they ‘luxuriate in the English word-garden’, as Simeon Potter put it. A familiar example is television, which (via French) yokes Greek tele– ‘far’ to Latin visio ‘seeing’. Neuroscience joins Greek neuro– ‘nerve’ to science, from Latin scientia ‘knowledge’. Other hybrids include automobile, hypercorrection, lovable, merriment, monolingual, sociology, and talkative.

Frankenstein’s monster reads a hybrid word and collapses in a daze; Dr F. flees in fright and disgust.
Purists used to complain about hybrids as if it were somehow unsavoury to fuse morphemes from different languages. Maybe this attitude owed something to a fastidious temperament and a bias for classical learning. Jan Freeman, writing about these Frankenwords, said that ‘usage gurus who could flaunt their Greek and Latin did, and those who couldn’t copied them.’
Eric Partridge, in Usage and Abusage, said neologisms should avoid ‘unseemly misalliance’ and pay heed to ‘etymological decency’. Ralcy Husted Bell called jeopardize ‘a monster’, which seems a bit harsh. These phrases give the impression that hybrids are malformed abominations, hideous chimeras to be shunned and disowned.
In their influential King’s English, the Fowler brothers object to amoral on the grounds that a– is Greek, moral is Latin, and it is ‘desirable that in making new words the two languages should not be mixed’. H. W. Fowler later compiled the following ‘ill-favoured list, of which all readers will condemn some, & some all’:
amoral, amusive, backwardation, bi-daily, bureaucracy, cablegram, climactic, coastal, colouration, dandiacal, floatation, funniment, gullible, impedance, pacifist, racial, sendee, speedometer
Several are so commonplace that it’s hard to imagine them bothering anyone; others never caught on. Often it seems to be the newness wherein lies the main trouble: rarely is there a problem with well-established hybrids. On this point, Robert Burchfield found that ‘the arguments apply only to words formed in the 19C. and 20C.’
Fowler believed that word-making,
like other manufactures, should be done by those who know how to do it; others should neither attempt it for themselves, nor assist the deplorable activities of amateurs by giving currency to fresh coinages before there has been time to test them.
But even if we were to deny ourselves the natural, playful urge to neologise, who would do the testing to which Fowler refers? An elite cadre of grammarians and grammaticasters, or the general population whose language it equally is? Again I find myself siding with Burchfield, in his New Fowler’s Modern English Usage:
Homogeneity of language origin comes low in [language users’] ranking of priorities; euphony, analogy, a sense of appropriateness, an instinctive belief that a word will settle in if there is a need for it and will disappear if there is not — these are the factors that operate when hybrids (like any other new words) are brought into the language.
This to me is a more sane and tolerant stance, free of purist dogma and control-freakery. Rejecting hybrids in English just because their parts’ ancient origins don’t match is pointless peevery. Bryan Garner, in his Dictionary of Modern American Usage, writes that nowadays ‘only a few Classics professors’ object to them. Let us be thankful for that.
My only regret is that hybrid is not a hybrid and so does not describe itself the way portmanteau does. But it’s probably too late to do anything about that.
.
Updates: Ben Zimmer has drawn my attention to a T-shirt with the text: ‘Polyamory is wrong! It is either multiamory or polyphilia but mixing Greek and Latin roots? Wrong!’
Ben quotes a similar joke from Tom Stoppard’s play The Invention of Love: ‘Homosexuality? What barbarity! It’s half Greek and half Latin!’