Crazy Actions and Choices MAD

2011·15.7 MB·English

by  Daniel Ellsberg

uykriday, May 4, 12
e-ll/Craay actions, expectations
{See my Notes on Turserko and Naftali, Sept. 19, 2011)

TAN (Fursenke and NaQali): 2115 Sov planners think they can use tae nucs to
complicate” a US landing on Cubs, without ese!

~-212: MinDef delegates to Pliyev authority to use sac aucs (oral delegation signed
off bty K on Sept. 7—after JFK warning!—and remains in force cll Oct, 22 (Gribkav)

There are gambles, there are long-sho's, ane there are crazy gambles an lang
shots. Itloaked crazy ta FxComm and others thal K would altemp! effort: because
an illegal and dangerous response by the US was likely

twas not “crazy” to suppose that JFK would not attack once the missiles
‘were operational with warheads (ané,ta0 late to blockade}.

Iewas not erazy (though mast Americans didn’t imagine it, akerwards) to
ccanjeclure, ar hope, Ural JFK would nat zeveal the presence of the missiles ithe
discovered them betore the election (betore they were all there and aperational, i
he could prevent a leak

Or that he could prevent a leak, for some days:

‘Though it was far from certain; a middling possibility, perhaps a long shat
(NOT impossible); K may have hoped for this or even counted on it (somewhat
unreasonably but not crazily), though there is np direct evidence 0” this. (My
conjecture during my study’)

Rul: K persisted even after the warnings by JFK on Sept. and 13. (He could have
reconsidered sending warheads atter Sept. 4; but then, ae mightas well have
brought back MBMS then. But there had been a BIG, unprecedented investment by
then; he goes ahead, on KT principles; rather than 4 sure (domestic) Ioss in
prestige, gamble... gand case of K/T gambling with catastroahe (my model} (can |
attribute this mavel to my crisis study? Itreally derives in part from VN, K/T's
work, and Challenger, all later than L964! Iya gaod case: K's fear of imminent lass
cof Cuba, and then of imminent loss of stature with exposure by drawing back from
‘huge investment vs. attempt to win, ai risk af disaste

[Like [PKs postponement ofhis willingness te trade missiles, gambling —for a day—
‘on possibility vf winniag (without probably realizing what risks he was taking of
global catastrophe by postponing: just as K was pastponing, on October 26-27
‘This was not only a case ofthe "Tait malaccompl” madel, for K. Both exhibited the
gambling-with-catastrophe madel, which so often (almost always?) precedes actual
Catastrophe! Though not in this case—one of them won, the other suffered 3
political disaster but neither, ulsimate glabal estastraphe—but the latter was mare
than possible, Itwas really a fairly likely outcome (not, perhaps, odds-on}), given the
two gambles (wita JFK in ignorance of key elements, not aware of the extreme
Fisks).

K's unique awareness--given his “crazy” choices of sending che nucs and warheads
at all, without informing JFK (where even a bluff, on hoch, would have been
safficient to win: this did not have to be beyond reasonable doulat!}—mace him the
candidate who in the erul-garn "Ia" to give in first

With JFK—not because he was cowardly but because he was reaspnable, nat mad—
US conventional superiority in the Carihbean and nuclear superiority globally
(real—except for Europe!) did not guarantee success, or avoidance af disaster, or
even make it very likely and unrisky (contrary to tae hawks, and many later
analysts).

It took luck (and Khrushchev's lack of madness, in the end, having made mad easier
choices: an¢ K's lac of control of Castro, who was “mad,” ina nat-unusyal way’ for
‘young male nationalists, a “suieide homber") far both to avoid catastrophe, and for
IPR win

(When disaster is “highly possible," even ifless likely than not, avoidance of it can
be said to be “lucky” even itt is more likely thar not, ar not unlikely)

seems to have fad na "exit" plan (who ever does?}; no plan B for what to do if J&K
annaunces the deslayment before itis operational, and threatens—or, inevitably, is
pressed (o threaten—attack and/or blockade. "What if scerecy fails’—two lang
befare election for JFK to keep the secret? Keating indicated this was happening by
August 10, (Perhaps JFK denials reassured K that either his own false assurances
were working, or that |F ith his implicit offer to postpane t
sevelation lll alter the election, aborting a “crisis" by mutual cooperatian (as an the
U2, hofare Powers),

was complying wi

Soe, to, warns and assures, to deter: yet he fallod ta prevent JFK from a
‘om:mitraent. (Both fall to dozer). Having oiled to deter JPK from the warnings, it
‘vas crazy to continue {he merely speeded up—to assure some missiles operational
before the election, to assure that JFK would nat reveal them? Qr, to assure no
wasion before election? That would bea reasonable concern, thaugh it's nat clear
lat the Sovs held it

Crazy unless: he took new steps to deter. He had a winning card: the tac
nnaes, with warheats; and the 40,000 Sov troops (when did they debark7) Te could
have revealed these as soon as they arrived —privately, to JFK, to allow him to climb
down Irom his eanimilment! Or publicly, o prevent pressure en [PK Ww carry Bul his
warning. And he could go to the Security Council immediately, himsel! (With
Indications of possible invasion af Cubal) (He's not asking them to intervene, but to
access lawful meassures of defense uf aa ally, and to resirain the US! He could
have preempted and reverseul the tables on Stevenson!

Instead, he counted (I think) on JFK's cooperative silence. But this was
unreasonable, after the October 14 U-2 flight! Or Keating's Oct. 10 speech which
asserted White House “knowledge.” ) He couldn'thave hoped alter that that [P
didn't know: or really, that he could keep it secret for several weeks! (HIF didnt use
‘thar week—or 12 days-- well, itat al! Ilis intelligence on the US was nat very good!
Hedidn’t pick up what the {IS media did! What would he, what shawl he, have done
Ihe hae

Al nll, twas a reckless gamble, bordering on a long shot—arguably crazy, given
the stakes (not just ot backdown—which he seems not to have cansidered!—but at
escalation},

Let me make thatt stronger: Jt was crazy (though it ould have worked,
‘against JFK: in fac, it came close, almost did work} to believe that (il was not "tao"
likely tat] an American president would not discover the maversent befare it was a
‘aitaccompli” and that he would not (be compelled to) take strongaction, however
illegal and dangerous.

1k was on the order of (and closely related to) the erazy estimate of K's
general that the missiles would not be discovered by overhead recor, wilh
camouflage, among "the palm trees” (Discovery was clase to certain, well before
the election: espec ally given K's unexplained decision not to use the SAMs against
the 1-2, and nol to delay installation ofthe missiles uneil the SAMs were
operations)

Lwantto introduce the camcept af "crazy, mae” choices—not all of which are meant
to look mad!—as a category in analysis. As distinct from merely inadvertent, or
poorly informed, or unreasoned, or noa-optimal choices. There is not only the

litical uses of (the appearance of} madness.” There is “ordinary maciness.” (Not
linical” in the individuals.)

Compare: CI, the Bay of Pigs: e expectation ofa general uprising in Cuba asa
result of the Bay of Pigs was crazy (for those who held it: different set, perhaps
Tram those who predicted i)

r of low-cost, quickly successful invasion of Cua (though this was before
VN). No plan for guerrilla war, ar sense of what that would mean (either in 1964 oF
1962!)
[CS hope or helief that JFK would use US forces rather than lase, despite bis
verbal commitment ta ther (see LB] in VN) was nol crazy: it was even likely: but

‘wrong, But that would have landed us in the soup (of VN) (although, on an island

near us}. JPK was not wrong, nor LB}, in resisting.

Likewise, the expectation of assassination of Castro as a useful elernent; or of
a fake attack on Guantanama (or a Northwoods-type attack on a US ship: was that
contemplated? Perhaps not, given the “provocation” expected of an impending
defeat of the US-supported attackers wilhaut US intervention (as planned for
Mongoose). Was not US intervention seen by the C14 as essential ta success, as by
EGLan Mongoose? JFX’s ruling that cut should have aborted the plan; but they
banked on his reversing himself: which he did not do ir VX, or, in tae end, in Cuba Tl
‘None of these were crazy expectations: but they were gambles, arguably long-shots,
and they all lst.

Je was the overall venture, relying an the frst two beliefs, that was crazy
(including the belief that it could be cavert, with the US hand hidder, or deniable).

General rule here: Something may look crazy; and probably is, if tt loolks that
vay; yet have calculations behind it, including some guesses ane reasoning that are
both esoteric and not entirely wrong or unrealistic, some unacknowledged motives
that are “human,” not unusual, some not even discreditable, others discred:table but
nol uncommon; yet to discover these (perhaps long alter the events) does not
necessarily mean that the effort was not crazy, alier all! Mighly reasoned and
motivated courses by generally reasonable people can be and oflen are, erazy! (See
escalation in Vietnam in 1965! And Nixon's strategy!)

‘This is not only for the public, voters, Congress to know, asa strong hypothesis. [tis
for the president to know, confronting allies and opponents! Not anly “terrorists,
Ofcourse, this possibility is asserted and exploited by those who want large
“deterrent, retaliatory" forces; and who also want to use them for extended
deterrence (counting on the opponents to know this about the US!)

Yer there are contradictions in their estimates: they assume that the opponent may
be mad—yet deterratle (ATW: We want to make FS an act of madness for bur
‘opponent; yet we need forees for these that presume considerable madness in the
‘opponent [a mere 20-40 million dead might not deter him: he's almost Hitler,
though Hitler probably could not have heen deterred at all). Yet (JCS, nar AjW}: we
‘wane him to believe us mad, in NATO (or with FU threats elsewhere), and for him
not to he mad enough to test that

Tam going to argue, of course, that the whole 67-year nuclear arms program has
been crazy, despite enormous internal calculation on bath sides (and hidden partial
premises that "make some sense,” considered by themselves], crazy in many ofits
premises and priorities leading to a continuously crazy result in terms of risks (both
possible outcomes and likelihoods and uncertainties}, ordinary madness ofan
‘extreme though not unprecedented sort, armed with unprecedented weapons
‘orotelling an unprecedented and final outcome

Ironically, some of the choices made by both JFK and K were less crazy than they
‘appeared to be at tae time (though some, unknown tll much later, were even
‘raver, especially by K)—but the predicament they led to, the gambles they were
making toward the end, were even worse, more reckless, more dangerous, than
‘either of them knew al the time, nor any ane else for 25 years (or later, with
sulamarines: ar now with my revelation om RFK),
Friday, May 4, 12
ll/Crazy actions, expectations
{See my Notes on Fursenko and Naftali, Sept. 18, 2011)

--FAN (Fursenko and Naftali: 211; Sov planners think they ean use tae nues to
“complicate” a USlanding or Cuba, without esc

212: MinDef delegates ta Pliyev authority to use tac nucs (oral delegation signed
olf bey K on Sept. 7—a‘ter JFK warning!—and remains in force UIT Oct. 22 (Gribkov)

There are gambles, there are long-shots, and there are crazy gambles and Jong-
| sss looked crazy o ExCommand others that Kwoafdatiempt effort becaise
an illegal and dangerous response by the US was likely

missiles

twas not “crazy” o suppose that JFK would not attack onc
‘were operational with warheads (and, too late to blockade)

(coneettasmot 2m (hgh mas Ameren imagine aera to
conjecture, or hope, thar {FX would not reveal the presence ofthe missles Fhe

| discovered them before the election (before they were all ere and operational), if

{ ne could prevent leak;

Or that he could prevent a leak, for some days;

‘Though it was far from certain; a middling possibility, perhaps a lon
(NOT impossible); K may have hoped for this or even counted on it (somewhat
unreasonably oul not crazily), though there is no direct evidence of this. (My
conjecture dutiag my study)

st K persisted even alter thevnrings hy JFK on Sot. 4nd 12. (He could have
reconsidered sending watheas afer Sep. bu then he igh as well have
brought hack MIMS then, Bat therehad been a BIG- unprecedented Investment by
then he ges ahead on K/T principles; rather than a site (domésti) loss in
[prestige gamble nA god case of KT gaming with etaszaphe (my mod) (can
[bute ths modelo my eri study Iereally derives in pa rom VN, K/Ts
ar, and Challenger al later than 1964! Irs a gond case: Ks ear of imminent loss
poe eae setarer gene etn om

(0x. Nhingetivestment vs. attempt Guida risk of disaster.
en
((Eikd JFK's postponement of his willingness to trade missiles, gambling—fora day
[Sees ft ig (without probably realizing what risks he was taking of .).-a
slobal catastrophe by postponing: just as K was postponing, on October 26-27, Zach
ker.

OA? (Ben Qe
‘This was nog only a case ofthe “fait malaccompli" model, for K. Both exhibited the

sambling-vith~catastrophe modo, which soften (almost aay?) precedes actual
| yfgtsstrapie! Thoaghtnotn this case—one of thers wan, the other sulfered 3

politcal disaster hut nether, ultimate global eatastrophe-—but the falter was more
than pessble. twas really fairy likely outcome (not, perhaps, acds-on) given the
two gambles (with [FK in ignorance of key elements, not aware of te extreme
risks) ~ Lee Ce.

Paw Orie
2 Wsunique awareness-siven his trazy" choices f sending the nus ané warbeads
Ze q Atall. without informing JFK (where even a bluff on both, would have been
SEL suiicent to win: this did not have te be beyond reasonadle doubt!) —made him tae
SSSED candidate who in the endgame "had? to give in first.

With JF—not because he was cowardly butbecause he was reasonable, not mad—
US conventional superiority in the Caribbean and suclear superiority globally

| ([rezl—oxcept for Furape!) did not guarantee success, ar avoidance of disaster, oF
‘even make it very likely and unrisky (contrary to the hawks, and many later
analysts).

Ie took mek (and Khrusitchev’s lack of madness, in the end, having made mad earlier
choices; and K’s lack of eantrol of Castra, who was “mac,” ina not-uousual
young male nationalists, "suicide bomber") for bath to avoid catas
JFK to win.

ye {When disaster is “highly possible," even if less likely than not, avoidance of it can
be sald to be “lucky” even if itis mare likely than nol, or aut walikely.)

K seems to have had no “exit” plan (@iho ever does?): no plan B for what to do if PK
announces the deployment before iis operational, and thraazens—-ar, inevitably, is
sete [pressed o threaten stack and/or boca, What if secrecy fals">-too long
re election for [PK to keep the secret? Keating indicated this was happening by
August 10, (Perhaps [PK conials reassured K that ether his own false assurances
4 { vere working, or that JFK was complying with his implicit offer to postpone the

revelation till after the election, aborting a “crisis” by mutual cooperation (as an the
U-2, before Powers)

So he, too, warns and assures, to deter: yet he Failed to preven

‘commitment, (Both fal todeter), Having failed ro deter JFK from the warnings, it
ray ta continue (be merely speeded up—to assure some missiles operational

© “before the election, to assure that JFK would aot reveal them? Or, to assure no
invasion before election? ‘That would be a xeasonable concern, though.isnot-clear
thatthe Sovs Held ays, ate gata 1

Crazy unless: he took new steps to deter. He hada winning card: the tac
jnucs, with warheads; and the 40,00 Sov troops (when did they debark?) Tle could
have revealed these as saon as Chey arrived—privately, to JFK, to allow him to climb
j down from his commitment! Qr publicly, to prevent pressure on JFK Wo carry out his

| warning, And he could go to the Security Council immediately, himself (With
indications of possible invasion of Cubal) (He's not asking them to intervene, but to
accept his lawful meassures of defense of an ally, and to restrain the US! He could
have preempted and reversed the tales on Stevenson!

> Instead, he counted (I think) on [PR's cooperative silence. Butthis was
unreasonable, after the October 14 U-2 flight! Or Keating's Oct. 10 speech which
asserted White House "knowledge." ) He couldn't have hoped after that that JFK ah 18 ©

ke didn’t know: or really, tat he could keep ir secret for several weeks! (HE didn't use
that week—or 12 days-- well, iat all! His intelligence on the US was not very good!
He didn't pick up what the US media did! What would he, what should he, have done
ifhe bad?

Allin all it yas a reckless gamble, bordering on along shot—arguably erazy, given
the stakes (not just of backdown—which he seems not have considered!—but of
esvalation). ~

Uncen peauues Mer coreeeaeaees
Let me make that stronger: It was crazy (though it could have worked,

against JFK: in fact, it came close, almost did work) to believe that (it was not “too”

[lily taat) an Americen presicent szuld not disenver the movement before was

& ov \ “fait accompli” and that he would nat (be compelled to) take strong action, howe

'nd dangerous. —

twas on the arder of (and closely related to) the erazy estimate of K’s
general that the missiles would notbe discovered by overhead recon, without
camouflage, among "the palm trees.” (Discovery was close to certain, well before
the election: especially given K's unexplained decision not to use the SAMS against
the U-2, and not to delay installation of the missiles until the SAMs were
operational},

| want to introduce the concept of “crazy, mad! choices—not all of which are meant
YC look mad—as a category in analysis, A distinct rom merely inadvertent or
poorly informed, or unreasoned, or non-optimal choices. There is not anly the

5" (Not

“political uses af (the appearance of] madness.” There is “or
“clinical” in the individuals.)

Compare: C1, che Ray of Pigs: the expectation ofa general uprising in Cuba asa
result af the Bay of Pigs was erazy (lor those who held. ita different set, perhaps,
rom tase who predicted it}

Or af low-cost, quickly successful invasion of Cuba (though this wes before
Vn), No plan for guerrilla war, or sense of what that would mean (either in 1961 or
19621)

vs fey)
I

Wr
JS hope or belief that JFK would use US forces rather than Jase, despite his

verbal commitment to them (See Li in VN) was not erazy: twas even lkelyrbut

wrong, But that would have landed us in the soup [of VN) (although, on an island

near us). JFK was not wrong, nor LB), in resisting,

Likewise, the expectation of assassination of Castro as a useful element; or of
a fake attack on Guantanamo (ora Northwoods-type attack on a US ship: was that
contemplated? Perhaps not, given the "provacation’ expected of an impending
feat of the US-supported attackers without US intervention (as planned for
Mongoose), Was not US intervention seen hy the CIA as essential to success, as by
EGL on Mongoose? |FK's ruling that out should have aborted the plan; but they
banked om his reversing himself: which he did not do in VN, or, in the endl, in Cubs tl.
None of chese were crazy expectations: but they were gambles, arguably lang-shot;
and they all lost. % 30

twas the overall venture, relying on the frst two beliefs, that was crazy
[including the belief that it could be covert, with the US hand hidden or deniable).

General rule here: Samething may look crazy, and probadly i,t looks that
‘ways yet have calculations behind it, including some guesses and reasoning that are
both esoteric and not entirely wrang or unrealistic, same unacknowledged! motives
‘hat aze “uma,” rot unusual, some not even discreitabe, others disreditable but
not uncommon; yet to discover these (perhaps long after the events) does not
necessarily mean tal the effort was noterary. alter all! Highly reasoned and
motivated courses by generally reasonable people can be and often are, erazy! (See
escalation in Viemam in 1968! And Nixon's strategy!) (arn,

This is not only tor the public, vaters, Congress to know, as a strong hypothesis. Its
for the president to know, confronting allies and opponents! Not only “terrorists.”
OF caurse, this possibility Is asserted and exploited by thse who want large
deterrent, retaliatory” forces; and who also want ta use them for extended
doterrence (counting on the apponents to know this about the US!)

Yet there are contradictions in their estimates: they assume that the opponent may
be mad—yet deterrable (AJW: We want to make FS an act of madness for our
‘opponent; yet we need forces for these that presume considerable madness in the
‘opponent (a mere 20-40 million dead might not deter him: he's almost Hitler,
though Hitler probably could nat ave been deterred at all). Yet (JCS, not AIW): we
‘want him to believe us mad, in NATO (or with FU threats elsewhere}, and for him
snot to be mad enough to test that

am going to argue, of course, that the whole 67-year nuclear arms program has
een crazy, despite enormous internal calculation on both sides (and hidden partial
premises that “make some sense,” considered hy themselves), crazy in many of ts
premises and priorities leading to a cantinuously crazy result in rms of risks (both
possible outcomes ant likelihoods and uncertainties), ordinary madness of an
extreme though not unprecedented sort, armed with unprecedented weapons
foretelling an unprecedented and final outcome.

Ironically, some of the choices made by both IFK and K were less crazy than they
‘appeared to be at the time (though some, unknown till much later, were even
crazier, especially by K)—but the predicament they led to, the gambles they were
making toward tne end, were even worse, more reckless, more dangerous, than
either of them knew atthe time, nor any one else for