Counting the days

I've argued before that I don't think the days of creation in Genesis 1 should be understood as calendar days (or solar days, human days, "normal" days, or whatever). The Hebrew word for day, yom, can be defined -- in fact can be literally defined -- as an extended period of indeterminate length, and if you think it should be understood metaphorically that opens up even more possible definitions.

One common objection to this is that when yom is modified by a number (e.g. first day, second day, etc.) in the Bible, its meaning is restricted to a calendar day. There are a few problems with this, but the biggest one is that it's false: yom plus a numerical modifier is used in the Old Testament to refer to a period of indefinite length. The best example of this is Zechariah 14:7-8 which uses the phrase yom echad (day one) to refer to a long period of time. Many translations do not translate that phrase as "one day" or "day one" but that is the Hebrew phrase. Here's the passage:

It will be a unique day [yom echad], without daytime [yom] or nighttime [layelah] -- a day known to the LORD. When evening ['ereb] comes, there will be light [or]. On that day [beyom] living water will flow out from Jerusalem, half to the eastern sea and half to the western sea, in summer and in winter. 

This verse tells us that there is a day known only to God in which there will be no daylight and no night, and which will encompass the annual seasons. As such, the day in question is an extended time period. The significance of this passage is threefold. First, obviously, it gives us an example of yom being used with a numerical modifier to refer to a long period of time. Second, the word yom is used twice in close proximity, but has two different definitions: daylight and an indefinite period of time. This is precisely what I'm claiming is the case in the account of the first day of creation in Genesis 1:5, which reads

God called the light [or] "day" [yom], and the darkness he called "night" [layelah]. And there was evening ['ereb], and there was morning -- the first day [yom echad].

If Zechariah 14:7-8 uses yom to refer to daylight, and yom echad to refer to an undefined period of time, there's nothing unusual in claiming that Genesis 1:5 has these same two definitions as well. In fact, these passages are the only two instances of "yom echad" in the entire Old Testament. This strongly suggests that the first day of creation was not a calendar day.

This leads to my third point: Zechariah 14:7-8 also contains several of the other terms in Genesis 1:5, such as or (light), layelah (night), and 'ereb (evening). That makes it the closest semantic parallel to Genesis 1:5 in the Bible. In order to defend the calendar-day interpretation, one would have to say that all of these parallels -- the same words, as well as yom echad referring to a long time period in close proximity to yom without modification referring to daylight -- are irrelevant. I'm afraid I don't find that position to be credible.

(cross-posted at Quodlibeta)