Let me say up front that I dislike the name "Shenandoah" for this song, but I'm using it because it's commonly recognized.

The curious-minded might have noticed that the METER of "Shenandoah" is mysterious, especially when considered in relation to the song's shipboard application, as a chanty.

Nearly every document of this song, in print representation, has a different meter. Often, it is a mixed meter, moving erratically between 3 beats and 4 beats.

This may be because 1) Singers in practice sang it with no consistent meter 2) Those who wrote it had trouble writing its irregular, yet consistent meter (I think we can rule this out at some leve) or 3) the song wasn't sung in strict meter at all.

I want to rule out #2 for now, having faith in at least a few authors' transcription abilities.

And I want to table #3. I speculate it likely WAS sung without strict meter at times. Recorded performances would give of evidence of this. Moreover, I know how to make songs without strict meter fit certain working tasks. Among these tasks are hauling halyards and rowing a boat. I have done both of these actions in real life while singing chanties without strict meter. Indeed, I have sung the St. Vincent/Grenadines "whalers" version of "Shenandoah" -- which I know from hearing the whalers sing it themselves -- while rowing a whaleboat.

The reason I want to table this #3 is because historical writers called 'Shenandoah" a windlass song. That doesn't mean it wasn't sung at other times (i.e. the Grenadine whalers rowed to their version) but that it was at least sung at the windlass. I'm assuming here the windlass means the lever windlass (which is pumped up and down) as opposed to the so-called windlass which is really a windlass connected to a capstan. At the moment, I don't have evidence organized to support that assumption-- I'm just going for it!

So, the windlass is an action that I believe (again, from my experience doing it on ships) requires a meter -- a regular beat. Beats fall in conjunction with actions of the workers continuously. Further, the operation of such windlasses entailed repeated cycles of 4 exertions. That is to say, a song verse would normally get chopped into four parts.

In an 8 measure song like "Sally Brown" (which was sung at the windlass, too). The first solo would cover one cycle of windlass action, the first chorus would cover another cycle, the second solo another cycle, and the second chorus another cycle.

SAL -ly BROWN was a CRE-ole LA-dy
WAY HEY ROLL and GO
SAL -ly BR-own was CRE-ole LA-dy
SPEND my MONEY on SAL-ly BROWN

The capitalized parts show the timing of 4 exertions per section corresponding to the rhythm of the melody.

The issue is that "Shenandoah" does not "fall" against the windlass action like this... and like nearly all other "windlass" chanties clearly do. This is curious.

Now, it may just be a fact to live with that "Shenandoah" does not line up neatly. Yet again, as someone who has sung chanties at windlasses, I personally find it jarring. One's body -- at least as I experience it -- gets used to the lines of a chanty falling at consistent times against the windlass' pumping action. That coordination, indeed, keeps everything flowing smoothly. "Shenandoah" may be an outlier, and I'll just have to live with that. It might mean that there is (was) no consistent way of singing "Shenandoah" at the windlass. But I want to suppose that there WAS, and get closer to answering "How?"

I believe the first reliable document of the melody of 'Shenandoah" is in Alden's _Harper's_ article of 1882. He actually give 2 different forms. The second of these has the longer second chorus with which most are familiar. What is remarkable in this (almost the first) printing of the tune is its absolute metrical regularity—in groupings of 3 beats. Subsequent documentations would be in the mixed (irregular meter) or in 4 beat groupings that, while viable, seem to me to distort the timing of the song.

Here I've done an absolutely strict, metered rendering of Alden's two forms - comparing them to an irregular form (mixed meter) that is familiar to many (coming from Terry 1921) and to the un-strict form of the Grenadine whalers.

https://youtu.be/V_CnZytGXUI

I will continue with theories, but I wonder what anyone thinks in the meantime.