There has been no smoking-gun account of how we got yet that reaches that far back. This has left space for athletic surmises such as that yet originated as a command "Get!", or that it began in Hebrew, where the equivalent happens to be the wanly similar 'od (upon which one commentator noted, "it must have had a long journey northwards").
This kind of thing makes us forgive those who have decided it is better to just classify yet's origin as "unknown," as does the dense thicket of more scientific but notready-for-prime-time speculations over the years. Liberman takes a machete to the overgrowth and reveals that yet is a fascinating shard of random accretions. Namely, of the three sounds that comprise yet, y is the spawn of a mistake; e is the shadow of what started as a whole word with a different meaning; and t is the remnant of a suffix now extinct.
Yet began as a word with two pieces: a word ei meaning there and a suffix -ta appended to it that meant roughly to. This ei-ta was in Proto-Germanic, the language that spawned English, German, Dutch, Icelandic, and Swedish. Ei-ta meant "there-towards."
In the Proto-Germanic branch Old English, ei-ta fused into a single word, ge¯t. The old - a ending was usually dropped these days. The g happened as a copycat phenomenon: at first, ei-ta had simply become e¯t. But there had been another short little adverb in Proto-Germanic that was used with a to-word appended to it. It was iu, which meant ever, hence iu-ta, "evertowards." So: to a Proto-Germanic speaker there were ei-ta and iu-ta, with meanings that felt similar.
Something happened with iu-ta. The i sound in iu started being pronounced as a y (yu-ta) because "ee" and the y sound are close in the mouth. Then this y sound gradually morphed into a g sound because the g isn't that far from a y in the mouth, such that by Old English, this iu-ta word was now gi¯et. Thus instead of the old ei-ta and iu-ta, there were now e¯t and gi¯et.
To Old English speakers, it felt natural to start pronouncing e¯t as ge¯t since there was that word with a similar meaning gi¯et, just as some people say yourn on the model of mine: humans have a natural drive to iron out a language's wrinkles.
If you are wondering why neither "there-towards" nor "ever-towards" sounds much like the meaning of yet, you are hardly alone: the problem threw etymologists for centuries. The "yet" meaning developed because Old English tended to use gi¯et in two set expressions. Nu¯ meant now, and pa¯ meant then, so nu¯ gi¯et meant "until now": that is, "now-ever-towards." Meanwhile pa¯ gi¯et meant "furthermore," where "ever-towards" lent a sense of movement beyond then-ness: "then and beyond."
After a while, when saying "until then" or "furthermore," people started leaving off nu¯ and pa¯ and just saying gi¯et. This is what languages do: in French, properly one negates a verb by placing ne before and pas afterward: Je ne parle pas "I do not speak." However, in spoken French, one usually leaves the ne out and uses only the pas "Je parle pas" which means that pas, which started as the word for "step," now carries the whole burden of negation.
In the same way, the elision of nu¯ and pa¯ left gi¯et carrying the meaning of both "until now" and "furthermore," which is, if you think about it, what modern yet means: He hasn't come yet (i.e., up to now); I'll show you yet (i.e., in the "furthermore)." Ge¯t, felt as a variant of gi¯et, took on this new meaning as well. By early Modern English, the g's had gone back to y's, and gi¯et was yit while ge¯t was yet: yet won out.






