Boats Against the Current

Posted by on

So we beat on, boats against the current, borne back ceaselessly into the past.

– F. Scott Fitzgerald, The Great Gatsby

George Cukor’s solid 1938 screwball comedy Holiday follows a classically satisfying three-act structure.

In Act I, Cukor reveals the problem facing our protagonist, Johnny Case (Cary Grant). Case is going to visit the home of his newly met newfound true love, only to find that she’s the daughter of staggering Park Avenue wealth. Hilarity ensues as Case meets Julia Seton’s (Doris Nolan) family (notably Katherine Hepburn as sister Linda) and the scale of their wealth is revealed room by Park Avenue mansion room. But as the first Act closes, we are introduced to the problem, as Julia breaks the news to her stuffy, wealthy father that she is engaged to be married. In classic screwball comedy fashion, this does not go well.

In Act II, the tension builds, as the Setons’ wealth and daddy’s expectations clash with Case’s values and desires for his life. The peril grows (it’s a screwball comedy, so the peril is of a certain sort) as we see how unhappy Julia’s sister Linda and brother Ned are, trapped by the wealth. (It’s fun in a Monopoly moneybags guy kinda way to see how Depression-era filmmakers treat gilded age wealth.) You see the tension building as the veil slowly lifts from Case’s eyes (Cary Grant is so good here.).

In Act III, the crisis comes to a head at the New Years Eve party/engagement announcement at the Seton mansion. Cue gowns, and tails, and lots of up and down the stairs. (“The third act must build, build, build in tempo and action until the last event,” to quote Billy Wilder.) Cukor’s good but not great here, because of the way he lets the energy flag a bit with a time gap before the crisis blows up. But then the resolution – Case walks out on Julia, but Linda follows, she’s really the girl for Case, which of course we saw coming because Hepburn got the film’s top billing alongside Grant. Satisfying ending, happily ever after, QED.

It’s not exactly classic – it was nominated for the Best Art Direction Oscar, who remembers that – but it’s a solid piece of work from the Columbia factory line. Cukor stuck with a reliable formula, the script is sharp and funny, and you can’t go wrong with Hepburn and Grant (and Edward Everett Horton, we made a note to run down as many of Horton’s movies as we can find for free or cheap).

on structure

I’m at that point in the new book project where I’m wrestling with structure. My co-author Bob Berrens and I have assembled a ton of material, driven by our shared love of deep and slightly chaotic research. (Slightly? Eye roll.)

My deep dive into classic Hollywood is mostly just shared fun (Lissa and I love watching old movies) but I can’t watch a good storyteller assemble a piece without thinking about how they did it. Bob and I have a growing pile of building blocks. How to assemble them?

There’s a scene in Cameron Crowe’s book of interviews with Billy Wilder, Conversations with Wilder where Wilder asks Crowe if he has an ending for the book. It’s part of the book’s meta-narrative, as Crowe or Wilder or sometimes both turn to discussing the making of a book in which they are the two lead characters. (It’s a great book.) Crowe hedges, and Wilder suggests the obvious solution. Wilder should die!

Of course this is real life, and Wilder doesn’t die. The book has a nice kicker, a hilarious Wilder quote, but there is no closure.

And that is the challenge of writing about rivers and people. There is no ending, no closure.

One of the core themes of the new book is the question of obligation – obligations to those upstream and downstream, of the past toward the future, of the obligations to the river itself. Our interest is people and the rivers with which they share the landscape, the manifold ways that relationship plays out. At the heart of that is the question of obligation. When people disconnect a river from its flood plain, they take on a forever obligation of maintaining the river in its new form. Or more generally, the forever obligation associated with the linkage of people and their water.

The river changes, people change, values change, institutions change. This, I have read, is the classic innovation of Shakespeare – the ideas that characters evolve over the course of the play.

This version of narrative structure works great in my stories about rivers. Until it doesn’t.

People embrace a river, but by the end of Act I it’s clear that the river is pushing back, revealing its true self. That’s the pre-story in Ribbons of Green as Puebloans, Spanish colonizers and early Americans settle in the Middle Rio Grande Valley.

Act II is filled with the struggles to respond to that riverine true self – the floods, the swamps, the droughts. The struggles build, the communities evolve in response. There are half-solutions, and change, and more half-solutions, and more change.

Act III is where the classic Hollywood three-act structure falls apart for me, because I can’t build toward a catastrophic moment that in a flash changes everything, what Wilder calls “the last event,” followed by a tidy tying up of loose ends.

This is where Wilder shines, because like F. Scott Fitzgerald’s famed kicker in The Great Gatsby, the great problems of their narratives, and mine, are never done.

How would Lubitsch do it?

Posted by on

Two literary frames of recent usefulness:

    Both are literary creations aimed at the widest possible audience – Holmes a fictional character created by Arthur Conan Doyle, Wilder a fictional character created by Billy Wilder.

    Holmes’ value is epistemic, Wilder’s (like my old standby “What would John McPhee do?”) involves narrative structure, arc.

    Wilder made a very bad but also very good movie about Holmes (even Wilder’s worst movies reward – the guy always did the work, which can be at its most interesting when the seams show – ditto Conan Doyle – the two frames are not unrelated).

    See also Lubitsch, Ernst.

    Tending our gardens, writ large

    Posted by on

    My office mate, pal, and podcast co-host Rin Tara and I switched things up for this week’s episode of our Water Matters! podcast, with Rin interviewing me about the Berrens-Fleck project Ribbons of Green. Among other things, Rin zeroed in on “gardens,” a theme that sits at the heart of the book. From the introduction:

    In Voltaire’s classic satirical eighteenth-century novel Candide, the title character and his fellow travelers settle into a simple division of duties after a chaotic life. They act collectively on a small plot of land, producing “plentiful crops” and pursuing a common goal. Candide’s final wisdom is famously and most often translated from the original French as follows: “Let us cultivate our garden.” But modern translations of Voltaire’s “Il faut cultiver notre jardin” suggest a broader interpretation, with the garden not as the plot of land behind the house but as a more expansive landscape: We must work our land. Translating jardin as fields or land points toward not just one’s personal space but also the larger community.

    We think of hoes, shovels, and rakes as the implements of our gardens, as our tools. But the expansive form of the noun—Candide’s fields, the Rio Grande Valley as our Kent, the grand Garden of New Mexico—calls for something more, a reliance on collective action. It is impractical for me to build a levee around my house to protect it from flooding or my own bridge to cross the river, and without the cooperation of my neighbors, I am powerless to drain a swamp encroaching on my back garden.

    Rin and I had lots of fun making the podcast, I hope you enjoy it.

    Quoting Sherlock Holmes

    Posted by on

    “Your fatal habit of looking at everything from the point of view of a story instead of as a scientific exercise has ruined what might have been an instructive and even classical series of demonstrations. You slur over work of the utmost finesse and delicacy in order to dwell upon sensational details which may excite, but cannot possibly instruct, the reader.”

    Sherlock Holmes in Doyle, A. C. (1904, September). The adventure of the Abbey Grange. The Strand Magazine, 28, 243–256.

    A grim Colorado River milestone

    Posted by on
    Graph showing total storage in Lake Mead and Lake Powell dropping below levels not seen since the 1950s.

    At Lake Mead and Lake Powell, a historic low

    On Sunday, the Colorado River passed a grim milestone. Total combined storage in Lake Mead and Lake Powell dropped to a historic low. It was my Wilburys colleague Jack Schmidt who noticed the milestone:

    On Sunday, July 12, the surface of Lake Powell was 3524.32 feet above sea level, and the surface of Lake Mead was at 1042.77 ft. These elevations equated to 5,505,869 and 7,169,640 acre feet (af), respectively, of live storage in the two reservoirs. The combined total live storage of these reservoirs was 12,675,509 af. The last time the combined total live storage in Lake Powell and Lake Mead was this small was May 23, 1957, during construction of Glen Canyon Dam when the entire amount of 12,668,000 af was stored in Lake Mead.

    Part of the subsequent analysis from the Wilburys team, building on Jack’s detailed reservoir storage analysis that’s been so central to our group’s work, is that it only goes down from here. For a while.

    Every day going forward, until runoff from the 2026-2027 winter snowpack begins next spring, a new record low will likely be set. This is a significant moment in the evolving Colorado River water supply crisis.

    Full writeup here.

    Cracks in the Upper Colorado River Basin wall

    Posted by on

    People are finally starting to say the quiet parts, the parts so many Upper Basin people have been grumbling about in private, out loud. Via an incredibly important Heather Sackett story:

    Colorado Sen. Dylan Roberts, a District 8 Democrat who represents several Western Slope counties, including Eagle, Grand, Garfield, Routt and Summit, asked Colorado’s lead negotiator, Becky Mitchell, whether the people of Colorado should have confidence that negotiations among the seven states that share the Colorado River have put the state in the best possible position. The states have been at an impasse for more than two years without a deal for future management as reservoirs continue to decline to record-low levels.

    “My constituents just see fighting and intransigence,” Roberts said. “And it’s concerning to me, especially as a Western Slope lawmaker … that the strategy is just ‘Let’s hire more lawyers; we’re going to court no matter what.’ That doesn’t give me confidence, because I don’t think Colorado fares well when we go to court against Arizona and California and Nevada, throwing our fate to the nine justices on the U.S. Supreme Court.”

    Recent Albuquerque monsoon history: dry

    Posted by on

    While pulling together some data today for our latest Water Matters podcast, I was surprised by the dry streak we’ve been in. The last above average monsoon here was 2018. (Episode posts tomorrow – 6/23 – I’ll try to remember to update the post with a link. Update: the episode)

    Also, I’ve been trying to up my data visualization game. For years I’ve had an R->ggplot work flow, doing the analysis in R and using it to generate most of the graphs I post on Inkstain as static image files. New LLM-based tools (I used Claude Code for this one) dramatically reduce the friction and make it possible (I originally wrote “make it easy,” but I’m still getting the hang of the form) to make them pointy-clicky.

    Feds stop paying to monitor Santa Fe drinking water source for Los Alamos contaminants

    Posted by on

    Via Alicia Inez Guzmán at Source New Mexico, we learn that the federal government is no longer paying the cost of monitoring Rio Grande flows at the intake to Santa Fe communities’ Buckman Direct Diversion as part of a joint effort to ensure that contamination from nuclear weapons work at Los Alamos National Laboratory doesn’t get into Santa Fe’s drinking water:

    the provision to pay for water sampling was “the stickiest point” of negotiating the MOU’s fifth renewal, said Justin Greene, a Santa Fe County commissioner and former Buckman board chair. The negotiations began the year before, he added, but the Energy Department’s rebuff hardened in 2025, the year the MOU expired. It was also the beginning of the second Trump presidency. 

    With the expiration, the rolling $96,000 that once funded annual stormwater sampling by an outside contractor was officially cut off and a new MOU was signed, effective in 2026, without it.

    For the Department of Energy’s nuclear weapons program, with a budget of something like $4b this year at Los Alamos, the $96k monitoring cost is change in among the couch cushions. I’d even characterize it as couch cushion change in the ~$300m DOE-lab budget for environmental cleanup of all the old messes left over from eight decades of nuclear weapons development.

    As friend of Inkstain Kyle Harwood noted, Santa Fe’s drinking water is safe:

    There is no risk to the drinking water quality, according to the Buckman’s consulting attorney Kyle Harwood, and the early notification system has continuously operated, even when the MOU expired. But until it can find another source of funding, the Buckman’s three partners plan to cover the costs of sampling for the lab’s contaminants through an ongoing cost-sharing agreement.

    It just means that the Santa Fe/Buckman ratepayers will have to come up with the money themselves, rather than the party responsible for the risk picking up the cost.