Coincidentally, today I looked up the etymology of “lintel” in the OED and was somewhat surprised to find that it came from an old French word for “threshold”, also that one of the meanings of “threshold” is given as “lintel”, although marked “erroneous”. It does seem strange and confusing that the same word can apply to both top and bottom of a doorway.
In my experience, people who don’t do carpentry are easily confused over terminology concerning carpentry, thus monks and other translators of Latin (not you Bayard) may not have had a good handle on what they were talking about here. For that matter, carpenters nowadays among the trades are the least conversant with the names for the parts, pieces, and things they work with. For example, there’s a term strongback, which is a beam that supports joists, typically in the middle of the span; there is also a perfectly good synonym from French which I can never remember, even though it’s equally common. Plumbers and electricians are very proficient with terminology. But, anyway, the AHD says lintel was an unattested alteration (by the French) of the Latin word limin or threshold. I don’t know much about Roman or ancient French architecture, but I think the idea was that with brickwork, absent a bricked arch, you put a cross-beam of wood over the doorway. Perhaps there was no really good name for such an item so they borrowed the word for the thing directly underneath the beam, that being the threshold. Apparently, in limine means “in the doorway” in Latin but the word literally means (modern English) threshold, the thing across the bottom of the doorway that is raised to some extent off the ground. The phrase ex eo superliminare must refer to the lintel (the header) or the part above the doorway which is in an opposite position from the limen or threshold. So in Latin, the word for lintel probably was superlimin, and perhaps the French just didn’t like saying it that way.
Unlike the Romans, the Anglo-Saxons engaged mainly in post-and-beam construction (all wood) with plaster in-fillings ("Tudor" style construction). The lintel or header is not a terribly important feature, or is a non-existant feature, in this scheme because the upper beam usually runs the entire length of the wall, or for as long as is feasible with the materials available. If the actual lintel is below the main beam (but above the door) it supports very little weight. In brick construction, by contrast, the lintel is short and used directly over the door to suport the brick, so it is truly a separate feature requiring a separate name. It also supports a great deal of weight, and much attention would be given to it’s construction. When the English got around to building things out of brick, I suppose they borrowed the pre-existing French term, lintel.
It strikes me as unlikely that the Anglo-Saxons were confused about the difference between a threshold, a doorway, and a lintel, but because Latin, and obviously the Bible, were the absolute authority, the translators faithfully translated the words as literally as possible. Thus, since apparently in Latin a threshold also meant a doorway, so too in Anglo-Saxon. Perhaps the Latin text refers to the blood in the doorway literally because the basin was considered to be in the doorway. Perhaps it is a mis-translation of the Hebrew (or Greek?). In any case, the purpose, as I understand it, of a threshold was to hold back the threshings because it was a pretty messy operation, yet you still needed a doorway to get into the room. Probably the actual, functional threshold was something high enough you had to step over it, say one to two feet high. But what was a threshold? The form þeorscwold must comport with German Tuerschwelle and maybe English speakers decided “thresh-hold” made more sense for a door-frame so took out the “w”. I don’t know. It’s possible that the “wold” portion of the word indicates a more primitive origin, perhaps a somewhat enclosed outdoor area. It would be an issue of technology just if, when, and how the operation went from outdoors to indoors. Wasn’t the early way of doing it a matter of threshing and then winnowing by the wind? In which case the “wold” portion was simply a flat area surrounded by mounded earth, that mound or berm being the referent in Dave’s first quotation above. Latin’s limin similarly moved from the meaning of a border between fields to the concept of a door sill and then apparently to the opening itself. Too many questions, not enough answers.
As an aside, subliminal means below the threshold of consciousness, so it still carries the sense of a barrier, as opposed to portal, which refers to the opening itself.
As an illustration of the difficulty with terms in the construction field, where workers are supposed to be more concerned with getting things done than talking about it, a co-worker refers to a shovel in Spanish as “una pala”. He also calls a cement trowel “una pala”, but if I don’t understand it he clarifies it as “una palita como esta” (with gestures).