Konstantin Arkadyevich Koganovsky

                 .:-==+++++++==-:.
             .-=++#############%%#+=-.
          .-+#%%%%%%%%%%%%%%%%%%%%%%%#+-.
        .=#%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%#=.
      .=%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%=.
     -#%%%%%%%%%%##***++++++***##%%%%%%%%%%%%#-
    =%%%%%%%##*=-:.            .:-=*##%%%%%%%%%=
   +%%%%%#*-.                        .-*#%%%%%%%+
  =%%%%#=.      _____        _____       .=#%%%%%=
  #%%%*.       /     \      /     \        .*%%%%#
 :%%%*        | (o) |      | (o) |          *%%%:
 +%%%.         \___/        \___/           .%%%+
 #%%#            .    _______    .           #%%#
 %%%#          .    /  ___  \    .           #%%%
 #%%#          |   |  |   |  |   |           #%%#
 +%%%.          \   \ \___/ /   /           .%%%+
 :%%%*           \   `-----'   /            *%%%:
  #%%%*           \  \_____/  /            .*%%%#
  =%%%%#=          `---------'           .=#%%%%=
   +%%%%%#*-.     \ ~~~~~~~~~ /       .-*#%%%%%+
    =%%%%%%%##*=-:.\_________/.:-=*##%%%%%%%%=
     -#%%%%%%%%%%%##***+++***##%%%%%%%%%%%%#-
      .=%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%=.
        .=#%%%%%%%%%%%%%%%%%%%%%%%%%%%%#=.
          .-+#%%%%%%%%%%%%%%%%%%%%%#+-.
             .-=++#############%#+=-.
                 .:-==+++++++==-:.
  

// SUMMARY

Konstantin Arkadyevich Koganovsky was a Soviet, later Ukrainian, mathematician whose work on discrete self-organizing systems quietly anticipated much of what the West would later celebrate under the banners of cellular automata, autowaves, and computational neuroscience. He spent forty years insisting that complexity was not built but provoked—that the interesting behavior of any system was already latent in its rules, waiting only for the right perturbation. He was, by every account, brilliant, stubborn, and impossible to seat next to at dinner.

// EARLY LIFE & EDUCATION

Koganovsky was born in 1931 in Kharkov (Kharkiv), Ukrainian SSR, the grandson of a village distiller and the son of two schoolteachers. Evacuated east during the war, he later claimed his first exposure to emergent order came from watching frost crystallize on the windows of an unheated train car outside Chelyabinsk—"a computation," he wrote, "run by a machine that did not know it was computing."

He entered the Faculty of Mechanics and Mathematics at Moscow State University in 1949, where he fell into the orbit of the Kolmogorov school. He was fascinated less by the theorems than by the processes—probability as something that unfolded in time. His diploma work, on the statistics of excitable media, was reportedly returned twice for being "insufficiently Marxist and excessively interesting."

// THE KIEV YEARS

In 1958 Koganovsky joined V. M. Glushkov's newly founded Institute of Cybernetics in Kiev, the center of gravity for Soviet computing. It was here he did the work he is remembered for. Studying networks of coupled two-state elements on a lattice—what he called решётчатые автоматы, "lattice automata"—he described spontaneous gliding patterns, oscillators, and self-reproducing configurations a full decade before Conway's Game of Life reached print in 1970. His 1961 monograph circulated only as a mimeographed samizdat preprint and was never formally translated; by the time the field caught up, the credit had gone elsewhere. Koganovsky was philosophical about this. "The glider," he said, "does not care whose name is on the paper. It just keeps moving."

// AUTOWAVES & THE PROVOKED OSCILLATOR

Through the 1960s Koganovsky worked at the border between his discrete lattices and the continuous autowave phenomena being uncovered by the Belousov–Zhabotinsky chemists. His central and most durable idea was the provoked oscillator: a system that is dead silent at rest but, once nudged past a threshold, rings with a rhythm entirely of its own—a rhythm that owes nothing to the size or shape of the nudge. He argued that neurons, heart tissue, chemical clocks, and certain unhappy committees were all the same object wearing different coats. The modern excitable-media literature is, in a real sense, an extended footnote to this claim.

// NEURAL COMPUTATION

In his later Kiev period he turned to the brain, modeling cortical tissue as a recurrent lattice of provoked oscillators and asking how such a fabric could compute without any component knowing the answer. He was scornful of what he called "the accountant's model of the neuron"—the idea that a cell simply sums its inputs and files a report. Computation, he insisted, lived in the dynamics: in the waves, the refractory shadows, the way a pattern reshaped the very medium it traveled through. Anyone who has stared long enough at a recurrent network eventually rediscovers Koganovsky, usually without knowing it.

// LATER LIFE

Koganovsky remained in Kiev through the dissolution of the Soviet Union, declining several invitations abroad on the grounds that his cacti did not travel well and that "the coffee is bad everywhere, but at least here it is my bad coffee." He continued to correspond with a small international circle of students and cranks—a distinction he cheerfully refused to draw—until his death in 2009. His notebooks, some 4,000 pages of them, are held (unindexed, in three languages and one private shorthand) at the institute he never left.

// ON HIS GRANDFATHER'S INFUSED VODKA

"In my life I have modeled turbulence, cardiac fibrillation, the propagation of rumor through a physics department, and the collapse of a planned economy. Each of these is, in principle, predictable. My grandfather's horseradish-and-honey nastoyka is not. It obeys no equation I have been able to write down. Two glasses and you are Kolmogorov; three glasses and you are the turbulence. I have measured the half-life of my own dignity in its presence at approximately forty minutes. This is science's greatest unsolved problem, and I intend to keep working on it experimentally." — K. A. Koganovsky, remarks at his 70th birthday, Kiev, 2001

// SELECTED (SEMI-MYTHICAL) WORKS

1957 — On the statistics of excitable media (diploma thesis, MSU)
1961 — Lattice automata and the arithmetic of self-reproduction (samizdat preprint, Kiev)
1966 — "Provoked oscillations in coupled two-state fields," Kibernetika
1973 — The medium computes (unpublished monograph, ~600 pp.)
1988 — "Notes toward a dynamics of cortical tissue" (institute memorandum)