Two weeks ago, I attempted to tie etymology to the most pressing problem of our time (see the post for 3 March 2021 on imps and vaccination). Now that woke and cancel culture have become the words of the day, I again saw my chance and decided to deal with the adverb yesterday. In yesterday, only yester– will interest us.
Not everybody may know that yesterday is one of the most enigmatic formations in the Indo-European language family. On the face of it, everything looks fine, because cognates are not wanting: Dutch gistern, German gestern, Latin heri “yesterday” ~ hesternus “yesterday’s, pertaining to yesterday” (the second Latin word needs no explanation; heri goes back to hesi), and many others. (English y- in yester– developed from g-.) Gothic, a fourth-century Germanic language, had gistra-dagis, an almost exact counterpart of yester-day. The Gothic Bible is a translation from Greek, but the Greek word gistradagis glossed was aûrion “tomorrow” (not “yesterday”!), and herein lies the main trouble.
The Gothic bishop Wulfila, a talented and most reliable translator, must have known very well what the Greek word meant. The context, with its juxtaposition, is also unambiguous: “…the grass of the field, which to day is, and to morrow is cast in the oven” (M VI: 30; the text and the spelling are as in the Revised Version). Even a non-specialist will notice the similarity between aûrion and Aurora, while those versed in historical linguistics will recall that English east is related to it (morning, or tomorrow, comes from the east). No doubt, gistra-dagis meant “tomorrow” in this Gothic verse.
How could it be? The plot thickens, for the Gothic case is not isolated. Those who are familiar with some modern Scandinavian language may remember Swedish i går “yesterday” or a similar form in Norwegian and Danish. The older version of this phrase also turned up in Old Icelandic, and it usually meant “yesterday” (as expected), but in a few isolated cases, including one in a well-known poem, the sense is “tomorrow.” The Icelandic phrase (í gær) had a long vowel in the root (the ligature æ always denoted a long vowel). Gær is related to gistra- (I’ll skip the details).
The ancient form gester– (as in English yester-) seems to contain the root ges– and the suffix –ter, with ges-, in turn, being made up of a prefix (g-) and some root like djes– “day.” By the time the word yester(day) and its cognates appeared in our manuscripts, its ancient inner structure must have been as obscure to the speakers as it is to us. The ancient phonetic shape of the initial prefix g- remains an object of rather fruitless speculation and need not interest us. We want to know how the same word could acquire two incompatible senses: “yesterday” and “tomorrow.”
The riddle has never been solved to everybody’s satisfaction. More than a hundred years ago, the great historical linguist Karl Brugmann suggested that the Germanic adverb meant “the day next to the present one,” with the context determining the reference. Perhaps his explanation, which is a restatement of the obvious fact rather than an explanation, can be improved. Didn’t the speakers of Old Germanic need more precise names for the day gone and the day to come? We may, I believe, approach the truth if we consider some of our ancestors’ views on time. Note that, unlike Greek and Latin, Germanic had no future tense. The present did all the work, as it still does in such English constructions as if you feel better tomorrow; when you return, give me a call, and I am leaving in two days. While reading a text in an old Germanic language, one often has to figure out whether the reference is to the present or the future. Prophecies obviously refer to the future, but not all situations are so clear.
In such matters, it is useful to step aside for a moment from the facts of language and take into consideration people’s oldest concepts of time. The ancient Greeks believed that the future was open to view (because it was in front of us), whereas the past remained hidden (no one can see what is behind). Another seemingly unexpected picture emerges from Old Russian chronicles. The Germanic noun kuningaz “king” made its way into Old Slavic and became knyaz’ in Russian, rendered in Latin as princeps and in English and German as prince ~ Prinz. Old Russian historical books mention numerous princes. Those who reigned in the distant past are called front princes (perednie knyaz’ia), and those who came after them are referred to as back princes (zadnie knyaz’ia). Shouldn’t it have been the other way around? Apparently, not: the rulers of the most distant past were at the front because everything began with them: the chronicler viewed history with the eyes of their characters, not with those of his contemporaries. Both examples show that the early ideas of time and its progress might differ from ours in a rather radical way.
I would like to suggest that for the oldest Germanic speakers, who did (and did very well!) without the future tense, the concept of and hence the word for tomorrow did not exist. The same was probably true of all the most ancient Indo-Europeans, who originally had no tenses but only grammatical aspects: they could describe the way this or that action was performed, divorced from the temporal framework. This approach is easy to understand. Both I have put the butter into the refrigerator, and I put the butter into the refrigerator refer to the past but characterize the past moment from a different perspective. In similar fashion, I speak and I am speaking refer to the present but also from different viewpoints. Given this picture of the world, adverbs of time are not always needed or, if they exist, their message is less precise than we expect. The Old Germanic languages developed the past tense and, predictably, an adverb like yesterday. Tomorrow and other references to the day after (and this is the core of my hypothesis) emerged later, perhaps much later.
If my suggestion is correct, for many centuries, yesterday (the same of course holds for its closest cognates) meant what they still mean to us and nothing else. Under the circumstances, it could occasionally become a default name for “any day next to the present one, an adjoining day.” Judging by the extreme rarity of the sense “tomorrow,” this usage struck the speakers of Germanic as inconvenient, perhaps even unnatural. Since in the extant Gothic text, gistradagis occurs only once, we cannot judge how common it was in Wulfila’s language. Not inconceivably, pressed for an exact gloss (the Greek adverb made him do so), he used the word for “the day passed” and assumed that his readers would understand the reference from the context. They, most probably, did. One wonders how Wulfila would have translated Macbeth’s tomorrow and tomorrow and tomorrow. (A postscript: yore in the phrase in days of yore is not related to yesterday.)
The title of this post (“the future in the past”) refers to constructions like “I knew that they would do well”: would, not will, because the future is viewed from the vantage point of the past. Odd, isn’t it? Our modern concepts of time and tense would probably have puzzled Wulfila no less than his gistradagis puzzles us.