JFH: A Scientific Obituary
Are you proposing that we eat my mother?
I am a bit peckish.
--“The Undertaker Sketch,” Monty Python's Flying Circus
Seven years ago this
spring, my older brother, Jessie Floyd Hayes, took his own life.
During his teen years,
when he still lived with us full-time on the farm, he was the “adult”
who paid more attention to me than any other. He introduced me to his
own obsessions: superhero comic books, serial killers, Monty Python,
Kolchak the Night Stalker, Art
Bell and the paranoid world of conspiracy theories. He
took me to my first theater movie, Short
Circuit, a very 80s comedy about a
conscious robot. Without his ongoing reinforcement, my weird
imagination might have withered, leaving me with not much more than a
good vocabulary and a memory for random facts. Of course, he was also
a socially frustrated teenage boy. Like all the other riders and
clowns in the testosterone rodeo, he was capable of petty cruelties.
Minor things, but they were what I chose to pay attention to, and
what informed my opinions of him at the time.
Later, during my
twenties, while I was studying up north, he would send me occasional
handwritten letters, saying he was both proud and jealous of me for
getting out, while he spent most of his life within fifty miles of
where he was born. Although he fantasized about joining a survivalist
community, and even drove to one of their conventions in Elko,
Nevada, he never really believed that he could break through to the
life he wanted. Eventually he gave up, and wrote out a will on a
pocket wirebound pad, and mixed several paper cups of death, which I
found, empty, in a neat line on a table across the room from the
recliner where he slept. I’m just now realizing that I have
never once looked at the state’s official autopsy/toxicology
report, so I don’t actually know what was in those cups.
And I’m not about to call my mom out of the blue and ask her.
Any discussion of the chemical aspect of his death will have to wait
for another time.
At the time of his
death, he was forty-six, the same age I am now. He was a
bachelor—overweight, balding, with thick glasses and teeth that
were crooked but otherwise perfect (never a cavity, his whole
life). He was born with no
sense of smell, probably due to a birthing injury that
sheared off the delicate olfactory nerves where they penetrate
the skull; this made shoveling cow manure on the farm
more tolerable, but may have predisposed him to depression.
He had recently been laid off, after fourteen years as a security
guard. He was still living independently, but he was near the end of
That spring, I was
developing a college elective on evolutionary biology for non-majors.
Buss's textbook, Evolutionary Psychology, I
found Catanzaro's theory
of suicide. It essentially says that childless, middle-aged males
kill themselves in unconscious recognition of their loserdom. Its
inherent Social Darwinism (losers deserve to lose) pissed me off so
much that I've never been able to let go of it. I don't deny the
statistical patterns that Catanzaro noticed, but I think a much more
nuanced picture is needed.
For instance, there's
good evidence that only some suicides are depressed. Others
are motivated by aggression,
as when this person fired a nail
gun into his head TWELVE times, and survived (a world
record at the time). In other words, people in an aggressive state
who don't have anyone else to lash out at will hurt
themselves. I personally saw this same kind of frustrated
behavior in caged lab monkeys while I was a graduate student. If they
got mad about something a handler, or another monkey, was doing, and
they couldn't reach the “villain,” they would bite
themselves, over and over. Not forever, not all day, but until they
had satisfied their urges to punish.
So I have a tendency to
complicate things. There was a passage in the book
Capture: Unraveling the Mystery of Mental Suffering
that grabbed my attention:
the one hand, the effort of the profession [psychiatry] has been to
make finer and finer diagnostic distinctions. We have the DSM-V,
which we use to diagnose and classify hundreds of different mental
disorders. And yet the biology is suggesting that these disorders are
related. It's no longer that schizophrenia is completely different
from depression—it looks very much as if there's a spectrum
that's interrelated rather than separate neurotransmitter system
diseases. And the drugs, wonderful as they may be, are not so
precisely targeted as is often claimed. We see one drug working for a
variety of problems—or not working. It seems to be often
largely a matter of individual response. There is also the growing
evidence of the huge placebo effect in psychoactive drugs, which is,
or ought to be, humbling.” (p17)
That book opens with an
ambition of finding underlying unifying mechanisms, but then spends
the next several hundred pages describing, through dozens of
case studies—which is exactly what psychiatry has been doing
for a hundred years. They're vivid, evocative descriptions—exactly
the kind of writing I'm always encouraging scientists to do to engage
the public with their work—so I do think the book is valuable
in that way. However, beyond mentioning the amygdala
and possibly misusing
the engineering term “feed-forward system” (when what I
think he means is a positive feedback system), there is no attempt to
develop a predictive model that could be falsified. In the
philosophical and logical traditions that serve as the historical
context for his “human mystery,” this is known as
If you sort through enough case studies, you can find some that will
support almost anything.
I've written before
about how mathematical modeling can put sharp teeth into a theory,
gnawing away those evocative but functionally unimportant details
until only the clean, predictive bones are left. There's a long
history of this kind of thing for simple physical systems, but less
so for complex systems like psychiatric
illness, where there are a lot more moving parts, and
you really need a computer to keep track of them. One of the great
modern success stories for modeling in the social sciences is
John Gottman's group. Their dynamical
models, consisting of what mathematicians call
Ordinary Differential Equations, have predicted whether couples in
counseling would divorce over 90% of the time, by focusing on
important details like the presence of contempt
and the ratio of failed connection
attempts made by each spouse.
The essence of
Kessler's idea in Capture is that mental illness is the
result of a positive
feedback loop, in which small momentary discomforts can sometimes
grow through reinforcement into long-term chronic conditions. Rather
than a negative feedback loop that reduces a disturbance over time
back to its target value
Target - Disturbance (a bunch of times) → 0
a positive feedback loop amplifies differences each time through the loop
Target + Disturbance (a bunch of times) → ∞
As Kessler points out,
this is not a new idea. William James said the same thing, using
words like “explosive.” So did the
Buddha, twenty-five centuries ago. The problem is that
words require interpretation. Ambiguity is part of their value.
Numbers, on the other hand, are valuable because they are inflexible.
When your theory is wrong, the discrepancy between your predictions
and the data will be a large number, and to make the number smaller
(hopefully zero), the predictions, and thus the theory, have to
change. Correctly predicting the likelihood of a particular person's
suicide in a specific time frame, for example, will require
What does any of this
have to do with writing science fiction? It might even seem
counter-productive, since evocative details and mystery probably sell
more books than clean predictive bones do. For one thing, I want to
see stories where psychiatrists have successfully embraced these
methods to better help patients. More generally, I think there's
narrative value in such a recursive
model of character development. We tend to focus on decisive,
dramatic events that flip a switch in a character's behavior (except
in horror and romance, where slow incremental changes build dramatic
tension). But that's not how the majority of humans actually develop
our personalities. According to George
Bonanno, only about a third of people exposed to
trauma go off the rails; two-thirds return to whatever their baseline
was. I hope to return to this issue sometime.
But for now, keep your
bag packed and your teeth clean. The end of the world
is ever nigh, and there will be no
fluoride in the water after the Collapse.
This column was
prompted by the anniversary of my brother Jessie’s death, but
also by the excellent obituary that my wife wrote for her own
brother, Ron Reall, who died suddenly of natural causes in December
TOTALLY going to see this in May. Updates later.
Losing smell may be more disruptive than never knowing it to begin with.
That perforated piece of bone is called the cribiform plate, and you can see it even in models of the skull.
There are multiple evolutionary theories of suicide, none of them particularly satisfying.
Check out the flow chart in Figure 1.
Of course, he was also high on meth, so that could have had something to do with it.
The math version. Fascinating but not easy reading.
A video compromise version, with the math partnership coming in at 11:00.
The non-math version.
GoogleBooks links are way too long. Scroll down; there are several good examples.
Cyclical Processes in Personality and Psychopathology
The paper itself is only three pages; well worth your time.
“Sampling bias is absolutely rampant in psychology.” Also, a nice follow-up on averages, after the January column.
“When a depressed person sees that loop emerge in her mind the urge to pull can be enormous. Agree with it and do something. Disagree with it and
prove it wrong. But either way, take it seriously. Pull, push; Click, click; automatic pilot mode.”
Read more by Randall Hayes