One thing that I remember being told by a relative old-timer who was in charge of the work shop at WheelSmith in Palo Alto was that my Stronglight 93 crank tapers should never be greased.
I wasn't even there that day for any crank-related reasons and had not mentioned the cranks, but as he looked over my bike that's what he decided to tell me.
I knew that I had oiled the spindle taper upon assembly but chose to remain quiet. He seemed to be speaking from many years of actual experience working on bikes with those cranks so I respected his opinion, but as a lightweight (who spread his mileage among several bikes) I wasn't much concerned about it. I've installed model93 cranks dry on the spindle ever since though.

This is a matter of worldview, I suppose. If you're more comfortable taking the word of one admittedly bright person over the broad consensus of thousands of other bright and professional persons, I'm not sure our discussion will be very fruitful. But let me reiterate that I am not at all interested in being right or wrong about this, because that's a false choice. I'm interested in a general rule, and the reasons for that general rule. Such rules allow for exceptions, since they prove (meaning test) the rule. And out of that testing comes improvement. Exceptions do not destroy the rule until they become more numerous.

In my view, it is important to be skeptical of received knowledge. It is also important to try to determine why that knowledge exists, and also how it is transmitted. I apologize in advance for this, but let's spend some more time critically analyzing the stories we have.

I'm not sure where you get your story from, but I think your description of the development process is largely inaccurate. I don't buy it. Bicycles were perhaps the single-most engineered consumer product of the 1890s and early 1900s in Britain, France, and the USA, and continued to be developed creatively for many decades to follow as new metallurgy and technology became available. It was very high-tech for its time; the bike industry wasn't just blacksmiths banging away in forges and chatting about it later in the village pub, this was huge companies with armies of engineers, and cities whose major industry was bicycles and parts. Many of those engineers migrated to the automotive and airplane industries, and brought what they learned about bikes into those industries. Others migrated the other direction, into the bike industry. There was always a lot of cross-breeding between bicycles and other engineering fields.
By the time square-taper cranks were invented in the 1930s, tapered press fits in machining and engineering had been around for almost 100 years. (To my knowledge, machinists recommend a dry taper for their tooling. But I'd suppose there's as much debate about this on their forums as we have about BB tapers.) Is it reasonable to assume that in its eighty-odd year history, not one of literally thousands of qualified engineers has questioned the "institutional wisdom" of dry tapers?
And is that, in fact, the institutional wisdom at all? That Brandt implied it was does not make it so. (He seemed to require something to topple in order to justify his attention to a subject.) It might be that it only appeared to become institutional wisdom because the reasons were not transmitted outside the community of engineers who design and work on them. All that came out was an instruction. For most people, that is enough. Most don't care why. They just want things to work.

Again, I would be very careful not to extrapolate Jan's experience very far. The most you can say at this point is that it applies only to his cranks. Period.

Regarding skepticism: In order for Brandt's account to be valid, We must take him at his word when he offers his account of why dry tapers were used. It seems to exist mainly as a rhetorical device to bolster his own argument. That's fine, but it's not engineering, and honestly, who knows whom he spoke to, or how many actual engineers he spoke to, or what their roles in the companies were? We just don't know that much. He could have embellished it to prove a point. I'm extremely skeptical of that sort of purely anecdotal account, especially from Brandt. He's at his best when providing empirical data and analysis and letting it stand on its own. Not so much when trying to prove a point in an internet argument.

Secondly, even if his story was true, what difference does it make? Maybe the people he talked to didn't know because they didn't need to know. There may be reasons he was not privy to, beyond either warranty concerns or mechanical perfection, that drive a decision.
And let me point out that that's what this is: a decision. There are consequences either way, greasing or not. That's the way it is with mechanical things. At some decision point, you weigh the consequences and make the call. Brandt's win-lose rhetoric notwithstanding, it's about understanding what happens either way

I don't doubt this phenomenon. I don't know if greased or dry tapers have anything to do with it directly. More on that later, I hope.

Again I apologize that my arguments so far have been light on engineering. One reason for that is that I am not a trained engineer (but my relatives were train engineers.) Another reason is that most of the people I discuss this topic with are not engineers either. So it usually boils down to "whom do you trust?" as I implied earlier. There are critical methods that can help us analyze that, and I think they can be brought to bear on this topic, as well as others.
So yeah, after having had these same discussions with mechanics and others for over 25 years, I'm weary of the method of argumentation that usually surrounds this. Brandt only fanned the flames of that when he weighed in on the matter. That flame caught on the tinder of the internet and hasn't burned out yet. So here we are.

I will say that I learned very early on in my career the "institutional wisdom" not to grease tapers. I received that and followed it, but not blindly, because there was always some debate about it. I have discussed the matter with knowledgeable people in- and outside the industry, and did my own research into the topic. I encountered exceptions to the rule. I have heard many arguments for and against it, and have tried to understand them on their merits, not having any particular personal stake in it beyond my reputation as a mechanic (which I take seriously.)

That's hilarious. The opposite conclusion. As the 93 cranks were really before my time I'd defer to the old timer. Most of my experience with them has been with my own bikes, not customer's. Any that I may have overhauled in the shop in the 80s would have gone back on no grease, as that was standard practice. ( I learned bike mechanics in the same area)

After some vigorous recollecting, I think I lightly greased mine because the cranks were old when I got them, and starting to get that tiny lip on them inside the crank arm tapers. I gave them a light swipe with a needle file and put the grease on it hoping to encourage reforming around the spindle without hanging up on the lip.

Really the worst outcome possible from greasing tapers is that a crank might creak. While that annoys customers, it isn't the end of the world. It's really odd that there's so much passion that goes into this subject.

Offhand, can't recall seeing a Shimano directive on this matter, as it pertains to square taper spindles. In the case their splined spindles, they are somewhat ambiguous, the wording implying that it's not necessary but is good practice.

On the other hand, Campagnolo documents from the same era specify thorough degreasing and even stipulate that the bolts must not be lubricated.

And to add another manufacturer, Phil Wood's instructions for installing cranks on their bottom bracket spindles include:
"To install your crank set, consult your original crank set’s installation instructions. A thin layer of grease should be applied to the spindle’s tapered surface before the crank arms are bolted on."

This is confusing and possibly contradictory: What are you supposed to infer if the original crank set's installation instructions specify a clean dry taper?

Personally, I'd absolutely give precedence to the crank manufacturer's instructions, since the spindle is much less affected either way than the crankarms, if indeed at all.

It says to consult the crank manufacturer's instructions - not necessarily to follow them blindly. Personally, I give preference to those who can give a coherent physical explanation of their viewpoint. That has led me to apply, or leave in place, a bit of lubricant at the spindle/crank interface.

It does not appear that you have read my previous posts. Please do if you wish to engage with me further in this discussion.

Never have I implied, much less stated, that one should follow manufacturer's instructions "blindly." You are arguing against something I never said, and if you continue to do so, I will ignore your posts. My point has consistently been that if you choose not to follow manufacturers' instructions, you had better know what the consequences are.
I gave you a coherent (if brief) physical explanation of my viewpoint. But you chose to ignore that. So again, why would it possibly matter to Phil Wood whether cranks from other manufacturers go on their spindles dry or lubed? Their spindles are many times harder than any aluminum crank, and as I said, won't be the least affected by lubrication or not. It's the crankarm that's going to be damaged or compromised by installation procedure. Not their spindle.

But anyway, there's no shortage of coherent physical explanations. How do you decide? A louder voice, or a more complicated or detailed explanation doesn't necessarily correlate to its being more authoritative. The basis of the explanation also matters.
These are a few that I happen to have instant access to. Submitted without commentary. If you have issues with any of them, or wish to know the basis of their explanations, I suggest you take it up with the sources themselves. They might actually respond.

Sutherland's: "The axle end and the hole in the crank must be clean and dry. Do not use oil, grease, or an anti-sieze compound. The tapered square system depends on the crank coming up firmly on the axle. Any lubrication will cause the arm to go on too far in tightening or to float on the spindle. Either way the arm will be ruined. Grease or anti-seize compound may be used on the threads of the crank bolt." (6th edition, rev. 1995, p. 2-4) [emphasis in original]

Barnett's: "...prepare the arm for installation by cleaning the mating surfaces in the spindle and arm with acetone or alcohol. The purpose of this is to remove any traces of lubricant. Since these two pieces are held together by friction, grease or oil may enable the arm to go on further (not necessarily a good idea.) Further is not more secure if arrived at by lubrication. Crank manufacturers are unanimous in recommending against spindle lubrication of the spindle when mounting the arm. Arguments to the contrary have been voiced, but never lubricate the spindle flats! If there is a concern about preventing corrosion or about contaminants getting in the gaps between the spindle flats and the hole flats in the arm, then treat the mating surfaces with Loctite 222 or 242.... The Loctite will seal the surfaces from moisture and dirt, reduce creaking problems, will not cause the arms to be less secure, and will allow arm removal with normal effort." (ch. 20, p. 11) [emphasis in original]

JIS: "Affirm that there is no adhesion of foreign matters and the like at the inner wall of corner hole of right crank and right corner face of bracket axle, and fit by coinciding the right crank corner hole with the bracket axle corner face." (JIS 1993, D 9311, table 4, panel 18)
The JIS spec is based on a test of the above assembly. It requires a load of 1200N being applied 1,000,000 times, 60 times per minute, (almost 278 continuous hours of testing) after which "...there shall not occur fissures or breakages in the crank, nor slackening in the fixing part of the crank and the crank spindle." (JIS 1993, D 9415, sec. 4-5)

All of the above recommend accurate torque on the crank bolt. Which (may I say again is Brandt's main point) is really the most important factor in the process. The missing point for most people is that lubrication on the spindle requires torque settings to be lower to compensate for the decrease in friction between the parts.

Very interesting. Makes me wish I had had access to a course on taper fit theory sometime in my education or career.

What's interesting to me is Sutherland's permitting of grease to be used on the bolt threads, but does not require this. This seems to me to be a huge variable left up to the installer, who has no test data to determine which is the more appropriate for use with the prescribed torque value.

Similarly, Barnett's allows the use of a liquid locking compound, which I am near certain would act like a lubricant during the press-fitting process. Again the installer is left to make the choice.

Is Campagnolo the only one who clearly specified the terms of assembly regarding both the bolt's threads and the spindle tapers?

Another curiosity is the torque specification for spindle nuts versus bolts. The nut's larger thread diameter and greater thread pitch both contribute to less axial force being generated in response to any given level of torque, but I don't recall that a different (greater) torque level is prescribed for them.
I always torque the nuts twice, before and after applying exaggerated rider weight to the pedals with the cranks held horizontal, since extra torque on these nuts often fractures the threads on the spindle. A greater level of press-fit is thus achieved without risk to the spindle's threads.

You can grease the taper however, if you do, you will have to lay the bike on its side, centering the crank arm over a tree stump, and give the opposing crank a wallop with a large piece of wood. Then... you can thread and torque the bolts in place to hold the cranks in place.