Gaps in the Record
10/8/94, Fairfax, VA, Patriot Center
In the course of live-tweeting this show, I made an embarrassing gaffe. Full disclosure: I typically stream shows off of PhishTracks or the PhishOD app, making a conscious sacrifice of quality for convenience. Usually, it works just fine — the point of this project is the music, not the fidelity. But for 10/8/94, I hit a snag when what I thought was a tape flip in the middle of Mike’s Song turned out to be a noteworthy edit, slicing out the one unique show note from the date: a mid-song “chant from a girls soccer team.”
According to the feedback, I didn’t miss much. But it was still a reminder that the historical record of Phish is incomplete. Part of why I started this project — aside from the fact that I like listening to Phish a lot — is that their career is documented with unusually high granularity for a musical artist, allowing for closer analysis of their evolution. To track the history of most bands, you’re stuck with their studio discography and perhaps a few notable tours or shows. With Phish, we have 1764 shows (and counting), almost-daily status updates covering large chunks of the last 32 years. It’s musical “big data,” to get obnoxiously buzzwordy.
And as with big data in the business or science world, it’s all too tempting to equate “big” with “complete.” The size of a dataset doesn’t have anything to do with its reliability; bias and errors can slip in just as easily and distort any conclusions hastily drawn. Missing out on a couple minutes of shrill teenager yelling might seem inconsequential, but if one of my working theories about what changed in 1994 is a more “serious” approach from the band, I could easily have chalked 10/8 up as hijinks-free supporting evidence, had I not peeked at Phish.net.
However, that’s a tiny and easily filled gap in the record. More opaque and more important to Phish’s development are the times when they’re not on tour, where our knowledge of their activities is scant. I mentioned a couple essays back the question of whether the band evolves more quickly between tours or on the road — an as-yet-unresolved query. But there’s no denying that some changes occur during their “offseason,” whether they practiced together every day, went off to work on solo projects, or just sat around the house and got a little bit older.
In data science, when you’re forced to deal with missing values, you often resort to interpolation; typically, averaging some number of data points before and after the gap to create a reasonable fill-in. Splitting the difference between the end of Summer ’94 and the beginning of Fall ’94, you catch a glimmer of what themes from the first half of the year carried over to the second. Last time we covered the first: a return to eccentric new material after supporting the more straightforward Hoist songs all spring/summer. The other two themes poke their heads up on this night just outside of DC.
One of these is a fundamental change in how setlists take shape from show to show. We’re still in the era where the band worked from pre-planned setlists, leaving little to chance. Over the course of listening to 1993, I grew frustrated with the formulaic approach, while remaining sympathetic to their desire to show off every facet of the band’s talent (and extensive wardrobe of gimmicks) on a nightly basis.
Early 1994 shows seemed to be squirming free from those presets. The biggest change is dropping the obligatory Big Ball Jam/Fish segment quagmire from late in the show, a jokey double feature that brought any second set momentum to a screeching halt. More subtle was the move away from dedicated jam slots, times approximately halfway through the first set and near the start of the second where the band reliably placed their biggest improvisational vehicles. If you wanted to make a bathroom run, it was a lot easier in those days.
We’re not to the golden age of the first set yet, but the second set is starting to show some welcome unpredictability. It showed up in the last two shows of summer, with their surprising Letter to Jimmy Page > Bowie and Antelope openers. Fall’s second sets aren’t as creative with their kick-offs so far — Maze and 2001 — but have dropped improvisational highlights later in the set than usually expected. In the final quarter of 10/7, where you’d previously expect the big balls to come out, they instead unleash a spirited Tweezer. And 10/8’s second set, after a stumbling start, hits a smooth stride for the remainder, from an eventful Mike’s Groove and a fiery Fluffhead to Purple Rain (the least intrusive Fish song) and a double-sided emotional peak of Hood and Suzy.
There have been second sets with strong flow before this, of course, but typically these were of the manic seguefest type, where the band is clearly off-script and freewheeling. 10/8 feels like it has more of a premeditated flow, where no one song is unexpected, they’re just all ordered to maximum effect. Looking back to summer, 7/16 II also comes off as a well-crafted, clearly orchestrated setlist, above and beyond the Harpua centerpiece, and that may have set a new standard for Trey’s pre-show preparations. Here also, an unappreciated assist can go to the now-customary appearance of Simple after Mike’s Song, extending the Groove to a quartet and thus filling out a hefty chunk of a set with a proven dynamic sequence.
The second theme apparently carrying over from the summer is a crucial aspect of Phish improvisation: patience. The most striking jam in the last show of summer, Antelope > Catapult > Antelope, took its sweet time getting through the song’s three sections, even as it grew in ferocity after Fish’s foot-in-mouth moment. On 10/8, we get the same restraint but in the opposite mood, with the latest in a long line of magnificent 94 Hoods.
In his 33–⅓ book on A Live One (which, fair warning, I will be quoting from incessantly as we stroll through fall 94), Wally Holland describes the Hood jam as following “the preacher’s dictum: Start low, go slow, get higher, catch on fire.” On display here is their ability to not just start low, but stay low, gently burbling for more than four minutes, barely tracing out the chord sequence, feinting at gaining in intensity but constantly pulling back, before finally, in a pillowy sort of tension/release, it launches full bore into its ascent. If Fall 1994 is known as a tour of jarring, restless jamming, Hood offers a powerful counterpoint — if the band wasn’t in a hurry, they could also produce magic.
Whether any of these objectives were consciously decided by the band during the black box between mid-July and early-October is anyone’s guess — interpolation is just approximation. Keeping those known unknowns in mind, as well as any unknown unknowns…such as edits masquerading as tape flips, is an important disclaimer in sorting out which Phish developments are happy accidents versus conscious strategy, as the changes we can actually hear start to happen at a quickened pace.