Ridgeline Pleasures

CP 1919 — Cambridge Pulsar, right ascension 19 hours 19 minutes — was the first pulsar ever found. The detection happened in November 1967 at the Mullard Radio Astronomy Observatory outside Cambridge, on a field of dipole antennas spread across a paddock roughly the area of fifty-seven tennis courts, built and run partly by hand. The pulses arrived once every 1.3373 seconds, sharper and more regular than anything astronomy had a name for. The formal modern designation is PSR B1919+21; the source lies in the constellation Vulpecula. Older accounts placed it about a thousand light-years off; more recent dispersion-measure estimates push it closer to two thousand three hundred. The star itself is indifferent to either figure.

For a few weeks, the discoverers labeled the signal LGM-1 — Little Green Men — because nothing natural was supposed to be that precise. Then more pulsars turned up. The mystery resolved into physics: a neutron star, the burnt-out core of a once-large sun, compressed to roughly the diameter of a city, spinning, with a magnetic field lashing radiation out from its poles in two narrow beams. One beam sweeps past the Earth on every rotation. We see a pulse. The pulsar itself is silent; the pulse is geometry.

The detection was made by Jocelyn Bell Burnell, then a graduate student under Antony Hewish, working through tens of feet of chart-recorder paper a night looking for scintillations. The 1974 Nobel Prize in Physics went to Hewish and to Martin Ryle. Bell Burnell was not included — an omission that has been written about, regretted, and refused to die ever since. She has, for her part, been gracious about it. In 2018 she received the Special Breakthrough Prize, three million dollars, and donated the entire sum to the Institute of Physics to fund the Bell Burnell Scholarship Fund — physics graduate studentships for women, ethnic minorities, refugees, and others underrepresented in the field.

Nine years after Craft's thesis, Bernard Sumner — guitarist of a young Manchester band called Joy Division — came across the figure reproduced as Figure 6.7, page 111, of the 1977 Cambridge Encyclopaedia of Astronomy, in Manchester Central Library. (Some accounts credit Stephen Morris with the find; Sumner is the more widely repeated version.) He brought it to Peter Saville, the in-house designer at Factory Records. Saville inverted the image — black-on-white became white-on-black — despite the band's stated preference for the original orientation, and used it, alone, on the front cover of their first album. No title. No band name. No text of any kind. Just the stacked waveform of a star nobody could see, pressed into textured paper, embossed, and shipped on 15 June 1979 as Factory catalog number FACT 10.

The decision to leave the front bare was as much the work as the image. The record had to be flipped to be named. The album itself was recorded over three weekends at Strawberry Studios in Stockport that April, produced by Martin Hannett, engineered by Chris Nagle.

Joy Division existed for four years. The lineup was Ian Curtis on voice, Bernard Sumner on guitar and synthesizer, Peter Hook on a bass played high enough to function as melody, and Stephen Morris on drums tuned and treated by their producer Martin Hannett in ways the band themselves found maddening and could not, after Hannett, replicate. They formed in 1976 in Salford as Warsaw, renamed Joy Division in early 1978, and released two studio albums — Unknown Pleasures in 1979, Closer the following year — with a small handful of singles in between.

Ian Curtis died by suicide on 18 May 1980, age 23, hours before the band were due to fly to America for their first US tour. He had epilepsy, a difficult marriage, and a depression that the medication of the time was not equipped to treat. The remaining three regrouped as New Order, added Gillian Gilbert on keyboards, and went on for forty more years.

It is not customary to discuss the cover without discussing him. The cover came first, by eleven months. He saw it. He approved it. It would later be very difficult, for everyone, to look at it without him.

If you or someone you know is struggling, the International Association for Suicide Prevention maintains a directory of crisis lines: iasp.info/resources/Crisis_Centres

The image has not stayed where it was put. It is now possibly the most reproduced piece of graphic design from the second half of the twentieth century — bootlegged onto t-shirts, embroidered onto Supreme drops, tattooed onto forearms, restitched as Mickey Mouse ears, pixelated into Stranger Things merch, painted on the side of buildings in São Paulo and Tokyo and Manchester. The waveform has become unstuck from its source — most of the people wearing it have never heard the album, and a smaller fraction still know it began as a chart of a neutron star.

In statistics, meanwhile, the technique itself acquired a new name. The phrase joy plot was coined by Jenny Bryan in a tweet on 24 April 2017; Claus Wilke shortly afterward built the R package ggjoy, then renamed it ggridges in September 2017 after sustained discussion of the source — Joy Division's name traces back to House of Dolls, Ka-tzetnik 135633's 1953 novella about Nazi concentration-camp sexual slavery (English translation 1955) — and the discomfort that name carries when transplanted to a statistical method. The naming is now split: joy plot in informal use, ridgeline plot in respectable journals. Both names trace back, through Saville, through Sumner, through a Cambridge graduate student counting pulses on chart paper, to a star that doesn't know what we did with its rhythm.

In 2015, Scientific American's Jen Christiansen tracked the image back to Craft, interviewed him at length, and published the full provenance in a feature that closed a forty-year loop on a figure most readers had only ever seen on a t-shirt.

Ridgeline Pleasures is the cover taken at its word — pulled off the flat, given depth, allowed to breathe. The ridgelines are not procedural. Each of the eighty lines you see is a real consecutive pulse from CP 1919, traced from Craft's Figure 5.37: three hundred samples per pulse, the same data the album cover used, brought back to three dimensions and given a translucent skin. Some pulses are tall, some are short, some doubled, some shifted in arrival time. The differences are the point. A pulsar is regular; an individual pulse is not.

A subtle breath modulates the amplitude over time, paced to 1.337 s — CP 1919's actual rotation period — not because the visuals are synchronized to the star, but because the wink had to be in there somewhere. The audio loop runs at half that — ~89.7 BPM, two beats per pulsar rotation — in A minor, with a sine kick, a noise-burst snare run through reverb in the Hannett tradition, a high melodic bass line in the manner of Peter Hook, and a slow filtered pad. None of it is a Joy Division song. All of it is from the same neighborhood.

Drag to orbit, all the way from looking down at the field to looking edge-on along it. Release and the drift resumes. The controls menu exposes amplitude, breath, rotation speed, line color, line thickness, baseline lift, row spacing, the pulse-range window (so you can isolate one ridge or one decade of the stack), and a camera-preset row for the canonical orientations.

The classroom uses for this piece are why it exists in interrogable form rather than as a flat poster. Five overlapping demonstrations get used in critique.

One: data and image are not the same thing. The same eighty pulses can read as a stack of waveforms, as a translucent landscape, as a wireframe terrain, as a contour study, or as eighty isolated curves. Toggle skin and lines, push the row spacing, and watch the design move while the data does not.

Two: the figure is the dataset. Pull the pulse range down to a single pulse and read it like a sentence — the burst, the noise floor, the asymmetric trailing edge. Sweep through eighty of those sentences and you have begun reading the paragraph.

Three: a design decision is reversible. Saville inverted Craft's white-on-black to make black-on-white black-on-white again. We invert it back to white-on-black to read the original. The piece lets you swing between them with one click. The line color slider lets the next student try amber, or pulsar-blue, or anything else they want to defend.

Four: a famous image is somebody's PhD figure. Tilting the field to ninety degrees lifts the data off the page into a thing you could print in resin and put on a shelf. The export button writes the same lines as an OBJ mesh and a Rhino-readable polyline set; what was once an album cover is also a model, also a curve collection, also a teaching prop you bring to your next crit.

Five: provenance survives translation. Every transformation in this page is undoable, and every claim about the work is sourced in the panels above. The point is not to keep the data pristine. The point is to make the chain of decisions visible: Bell Burnell → Craft → the encyclopedia plate → Sumner → Saville → Bryan → Wilke → you.