In the
United States Court of Appeals
For the Seventh Circuit
Nos. 08-3528, 09-1529, 09-1631
U NITED S TATES OF A MERICA,
Plaintiff-Appellee,
v.
L UIS G ONZALEZ, M IGUEL A YALA, and
F IDEL S. H ERNANDEZ,
Defendants-Appellants.
Appeals from the United States District Court
for the Northern District of Illinois, Eastern Division.
No. 04 C 1023—John W. Darrah, Judge.
A RGUED A PRIL 14, 2010—D ECIDED JUNE 15, 2010
Before P OSNER, R IPPLE, and K ANNE, Circuit Judges.
P OSNER, Circuit Judge. The three defendants pleaded
guilty to conspiring to distribute more than 50 grams of
mixtures containing cocaine base. Gonzalez and Ayala
were sentenced to 120 months in prison and Hernandez
to 300 months. Ayala’s appeal need not be dis-
cussed separately because it presents the same issues
as Gonzalez’s; Ayala’s brief incorporates Gonzalez’s by
reference.
2 Nos. 08-3528, 09-1529, 09-1631
Common to Gonzalez’s and Ayala’s appeals is the
recurrent issue—which should have been laid to rest
long ago—of distinguishing crack cocaine from other
forms of cocaine base. Cocaine is the active ingredient
of the coca leaf. After extraction from the leaf it is pro-
cessed into cocaine hydrochloride for export to the
United States and other countries by the addition
of hydrochloric acid and other chemicals. Cocaine hydro-
chloride is a powder usually consumed either by being
sniffed or by being dissolved in water and then injected.
(It cannot be smoked because heating causes it to burn
rather than vaporize.) Although cocaine hydrochloride
is a salt (a technical term in chemistry), it can be con-
verted to a base by various methods, two of which have
been involved in cocaine usage in the United States. The
first adds water, and a chemical such as ammonia
that removes the hydrochloric acid, and a further
chemical, usually ether, to separate the cocaine from
any remaining impurities. Pure cocaine is a base, and
the cocaine base produced by the method just described
is what is called “freebase.”
The second method involves dissolving cocaine hydro-
chloride in a solution of sodium bicarbonate (a
“weak base,” in chemistry jargon) and water, and boiling
the new solution that results. The base produced by
this method is called “crack.” The cocaine base produced
by either method is a hard crystalline substance that
when heated vaporizes. The inhalation of the vapor
produces a more rapid and intense intoxication than
sniffing powder cocaine or even (perhaps) injecting a
liquid solution of it.
Nos. 08-3528, 09-1529, 09-1631 3
Freebase was the first form of cocaine base to
become popular, but it fell out of favor because it was
difficult to make and because the flammability of the
ether created a risk of a serious burn injury, as famously
befell the comedian Richard Pryor. In the United States
today, though there are still freebasers, see, e.g., “Making
Cocaine Freebase With Ammonia Methods,” www.drugs-
forum .com /forum /show thread.php?t=30174&p age=2
(visited May 21, 2010), crack is generally believed to be
the only form of cocaine base that is widely con-
sumed. United States v. Plummer, 581 F.3d 484, 489 (7th Cir.
2009); United States v. Kelly, 519 F.3d 355, 364-65 (7th Cir.
2008); United States v. Griffin, 173 Fed. App’x 506, 509
(7th Cir. 2006) (per curiam); United States v. Brigman, 350
F.3d 310, 314-15 and n. 4 (3d Cir. 2003).
Congress has prescribed a mandatory minimum sen-
tence of ten years for crimes involving 50 grams or
more of a mixture containing “cocaine base,” 21 U.S.C.
§ 841(b)(1)(A)(iii), but has not defined the term. The
sentencing guidelines also prescribe increased penalties,
in the form of higher base-offense levels, for crimes in-
volving “cocaine base,” U.S.S.G. § 2D1.1(c)—but define
“cocaine base” as “crack.” Id., Application Note D. In a
series of cases culminating in United States v. Edwards,
397 F.3d 570 (7th Cir. 2005), this court has held that the
statutory term “cocaine base” likewise means “crack”—“the
street name for a form of cocaine base, usually prepared
by processing cocaine hydrochloride and sodium bicar-
bonate, and usually appearing in a lumpy, rocklike form.”
Id. at 572, quoting Application Note D, supra. One
reason for our adopting this definition was that pure
cocaine—the cocaine in the coca leaf, before any pro-
4 Nos. 08-3528, 09-1529, 09-1631
cessing—is a base, yet no reason has ever been sug-
gested why Congress would have wanted crimes
involving unprocessed cocaine to be punished more
heavily than crimes involving cocaine hydrochloride.
We thought it clear that “Congress intended the
enhanced penalties [for crimes involving cocaine base] to
apply to crack cocaine and the lesser penalties to apply
to all other forms of cocaine.” Id. at 574, quoting United
States v. Booker, 70 F.3d 488, 494 (7th Cir. 1995). For evi-
dence that that was indeed Congress’s intention see
The Crack Cocaine Crisis, Joint Hearing Before the Select
Committee on Narcotics Abuse and Control and the Select
Committee on Children, Youth, and Families, House of Rep-
resentatives, 99th Cong., 2d Sess. 26 (July 15, 1986); United
States v. Bryant, 557 F.3d 489, 500 n. 15 (7th Cir. 2009);
United States v. Booker, supra, 70 F.3d at 493-94.
The plea agreements in this case authorized a sen-
tencing enhancement for cocaine base only if the district
judge found that the substance sold by the defendants
“was cocaine base in the specific form for which en-
hanced penalties are required as set forth in” the Edwards
case. The defendants interpret this to mean that the
sentencing judge had to find that they had sold crack
that had been “produced by mixing cocaine hydrochloride
with baking soda and water, boiling the mixture until
only a solid substance is left, and allowing it to dry,
resulting in a rocklike substance.” This is a quotation
from Edwards (397 F.3d at 574) in which the qualifica-
tion “usually” prepared that way and “usually” rock-
like, which appears in the earlier definition of crack in
Edwards (and is also the definition in the guidelines),
was omitted.
Nos. 08-3528, 09-1529, 09-1631 5
There was abundant evidence that the cocaine sold
by the defendants was crack but little evidence con-
cerning how it had been produced. The usual method is
indeed by heating a solution of cocaine hydrochloride and
baking soda, the common name for sodium bicarbonate,
a weak base as we said. But crack can also, as remarked
in several cases, see, e.g., United States v. Bryant, supra,
557 F.3d at 498-500 and n. 9; United States v. Abdul,
122 F.3d 477, 479 (7th Cir. 1997); United States v. Waters,
313 F.3d 151 (3d Cir. 2002), be produced by heating solu-
tions of cocaine hydrochloride and other weak bases
that, like sodium bicarbonate, strip the chlorine and
hydrogen atoms from the cocaine molecule, yielding
water, a chloride compound, cocaine base, and whatever
is left of the weak base used to do the stripping.
It is a misreading of Edwards to suppose that the
identity of the weak base used to produce crack was an
element of our definition of the word. No one suggests
that the precise choice of the weak base is material to
the intoxication produced by crack, or to any other
aspect of the drug that is perceptible to a consumer
or relevant to Congress’s decision to punish crimes in-
volving crack more heavily than ones involving other
forms of cocaine. Different processes can create the
same product: water can be heated on the burner of a
stove or in a microwave oven; different software can
generate the same images on a computer screen. The
defendants’ insistence that Congress, and this court in
Edwards, were concerned not with the end product of
creating crack but with the particular weak base
normally used to transform cocaine hydrochloride into
crack is relevant to no conceivable penological concern.
6 Nos. 08-3528, 09-1529, 09-1631
The reason the issue of whether the cocaine base
sold or possessed by a defendant is crack keeps being
raised (as indeed it does, see, e.g., United States v. Plummer,
supra, 581 F.3d at 488-89; United States v. Betts, 576
F.3d 738, 744-45 (7th Cir. 2009); United States v. Brisbane,
367 F.3d 910, 912-14 (D.C. Cir. 2004)) is dissatisfaction
with the “street” definition of crack (“a form of cocaine
base, usually prepared by processing cocaine hydrochlo-
ride and sodium bicarbonate, and usually appearing in
a lumpy, rocklike form”). Something more technical,
more rigorous, more precise is desired but isn’t easy
to come up with, because all cocaine base, unless it con-
tains impurities, has the identical chemical composition.
That makes it difficult for chemists to determine
whether a given substance is crack, freebase, or some
other form of cocaine base, though the different forms
might be distinguishable on the basis of physical differ-
ences between cocaine crystals distilled out of ether
(the freebase method) or left after boiling a solution of
cocaine hydrochloride and a weak base (the crack method).
And thus in the present case as in most cases some and
often all the evidence concerning the drug possessed or
sold by the defendant is verbal (for example, “I saw
Gonzalez sell crack”) rather than the drug itself. The
defendants complain that the witnesses used the word
crack “colloquially,” but crack has only a colloquial
meaning; there is no chemical name for it because all
forms of cocaine base are chemically the same.
We are hard pressed to understand why after all these
years the Justice Department has yet to commission an
expert study of the drug trade that would confirm, what
Nos. 08-3528, 09-1529, 09-1631 7
is widely believed, that almost all cocaine base sold in
this country is crack. We say “sold” rather than “con-
sumed” because Internet sources indicate that some
drug users who want the faster, more intense high pro-
duced by cocaine base but don’t want to buy street-
quality crack are continuing to make and consume their
own freebase. See, e.g., “Making Cocaine Freebase With
Ammonia Methods,” supra. Even when freebasing was
popular, users purchased powder cocaine and prepared
the freebase themselves, using kits that contained ether,
instructions, and the necessary equipment, rather than
buying the freebase. Edith Fairman Cooper, The Emergence
of Crack Cocaine Abuse 88-89 (2002); Dorothy K. Hatsukami
& Marian W. Fischman, “Crack Cocaine and Cocaine
Hydrochloride: Are the Differences Myth or Reality?”
275 Journal of the American Medical Association
1580 (1996); “Melting Down,” Time (Apr. 11, 1983)
www.time.com/time/magazine/article/0,9171,923501,00.
html (visited May 21, 2010); 21 U.S.C. § 863(d)(15).
If the rarity of the sale in the United States of any
cocaine base other than crack (not only freebase but also
coca paste, which is smoked in South America, United
States v. Kelly, supra, 519 F.3d at 365) were confirmed
statistically (we haven’t found any statistics on sales
of the different forms of cocaine), there would be minimal
danger that a witness was using the word “crack” to refer
to some other form of cocaine base, or that a dealer in
cocaine base was dealing in something other than crack.
A study such as we have suggested would go far to
dissolve issues concerning the nature of the product
in which a defendant charged with a crack offense
8 Nos. 08-3528, 09-1529, 09-1631
dealt. Still, there is sufficient expert evidence, albeit
qualitative rather than quantitative, that the sale of any
form of cocaine base other than crack is rare, United
States v. Plummer, supra, 581 F.3d at 489; United States
v. Kelly, supra, 519 F.3d at 364-65; United States v. Griffin,
supra, 173 Fed. App’x at 509; United States v. Brigman,
supra, 350 F.3d at 314-15 and n. 4, to allay, in conjunc-
tion with the evidence in this case (and bearing in mind
that the burden of proof on the government in a sen-
tencing hearing is merely the preponderance standard, not
proof beyond a reasonable doubt, and that the rules of
evidence do not apply to sentencing hearings), any con-
cerns about Gonzalez’s and Ayala’s sentences.
Defendant Hernandez raises four issues. One is
whether Rule 410 of the federal evidence rules was
violated by the introduction at the sentencing hearing,
as evidence of drug quantity, of statements that
Hernandez had made in the course of plea bargaining.
Admissions generally are admissible—and very impor-
tant—evidence; and Rule 410(4) (which Fed. R. Crim. P.
11(f) applies to “the admissibility or inadmissibility of a
plea, a plea discussion, and any related statement”)
makes admissions in plea bargaining inadmissible only
if the plea bargaining either does “not result in a plea
of guilty” or results “in a plea of guilty later withdrawn.”
Neither condition is satisfied here. United States v. Paden,
908 F.2d 1229, 1234-35 (5th Cir. 1990); 1 McCormick on
Evidence § 160 (6th ed. 2009). Statements in plea discus-
sions that result in a guilty plea that is not withdrawn
have to be admissible—they are the basis on which the
judge decides whether to accept the plea and are also
Nos. 08-3528, 09-1529, 09-1631 9
an important basis for deciding on the length of the
sentence.
Anyway the rules of evidence, other than those that
enforce privileges, are inapplicable to sentencing. Fed. R.
Evid. 1101(d)(3); United States v. Atkin, 29 F.3d 267, 268
(7th Cir. 1994). “[T]he only requirement is that the
evidence supporting the sentence must be reliable.” United
States v. Cooper, 591 F.3d 582, 591 (7th Cir. 2010). That
requirement was satisfied by Hernandez’s admissions.
Hernandez was given an enhanced sentence for pos-
sessing a gun (actually two guns) in connection with
a drug offense. U.S.S.G. § 2D1.1(b)(1); United States v.
Are, 590 F.3d 499, 526 (7th Cir. 2009). He had been
stopped by the police while driving with his girlfriend,
and the police had found crack in her possession.
Hernandez made a deal with the arresting officers:
they would let her go for now in exchange for his sur-
rendering two guns to them. He got two guns from
fellow members of his drug gang and gave them to the
police as he had promised. Obviously he possessed the
guns when he gave them to the police, and the drug
conspiracy of which he was convicted continued into
the period in which he delivered the guns and he
admitted that the girlfriend was a member of the con-
spiracy. So the guns were possessed by him in connec-
tion with a conspiracy (as in United States v. Acosta, 534
F.3d 574, 588 (7th Cir. 2008)), the connection being estab-
lished by the fact that he used them to obtain at least
temporary freedom for a coconspirator. Police are eager
to take criminals’ guns out of circulation, see United
10 Nos. 08-3528, 09-1529, 09-1631
States v. Foster, 166 Fed. App’x 13, 18 (4th Cir. 2006) (per
curiam); Daniel D. Polsby, “Firearms Costs, Firearms
Benefits and the Limits of Knowledge,” 86 J. Crim. L. &
Criminology 207, 219 (1995), and this makes guns a form
of currency by which drug dealers and other criminals
can obtain concessions from the police that help them
stay in business. An FBI agent testified at Hernandez’s
sentencing hearing that it is the “routine operation or
practice of the Chicago police officers, beat officers on
the street in uniform, [to stop] people and hav[e]
them obtain firearms in exchange for negotiated deci-
sions not to arrest, or reduced charges or such things
as that.”
Hernandez complains about a further sentencing en-
hancement that he received, for obstruction of justice.
U.S.S.G § 3C1.1. After being arrested and told that he
would be prosecuted for drug offenses, he agreed to
cooperate with the government and on the basis of that
agreement was released from jail. Part of the agree-
ment was that he would keep in touch with the prosecu-
tors. But after a month he moved to Florida without
telling them and was there for two years before being
apprehended and brought back to Chicago to stand trial.
He argues that he was merely fleeing, and points out
that flight as such is not a ground for the enhancement.
U.S.S.G. § 3C1.1, Application Note 5(d); United States v.
Hagan, 913 F.2d 1278, 1284-85 (7th Cir. 1990); United States
v. Burton, 933 F.2d 916 (11th Cir. 1991) (per curiam); United
States v. Garcia, 909 F.2d 389, 392 (9th Cir. 1990); United
States v. Stroud, 893 F.2d 504, 507-08 (2d Cir. 1990). Instead
Nos. 08-3528, 09-1529, 09-1631 11
the cases suggest a distinction between “panicked” or
“instinctual” flight, “mere flight in the immediate after-
math of a crime,” or “spontaneous” flight (“the instinctive
flight of a criminal about to be caught by the law,” as we
put it in United States v. Hagan, supra, 913 F.2d at 1285), on
the one hand, and “calculated evasion,” or “a deliberate
pre- or post-arrest attempt to frustrate or impede
an ongoing criminal investigation, as in the case of an
individual who flees while being sought for ques-
tioning some time after the commission of a crime,” on
the other hand, and say that only the latter is obstruction
of justice. See, besides the cases just cited, United States
v. Arceo, 535 F.3d 679, 687 (7th Cir. 2008); United States v.
Porter, 145 F.3d 897, 901-04 (7th Cir. 1998); United States
v. Draves, 103 F.3d 1328, 1336-38 (7th Cir. 1997); United
States v. Walcott, 61 F.3d 635, 639 (8th Cir. 1995); United
States v. Madera-Gallegos, 945 F.2d 264, 266-68 (9th
Cir. 1991).
But the adjectives obscure a simpler distinction.
While anything a criminal does to avoid being caught
increases the burden on law enforcement and so could
be thought an obstruction of justice, to reason thus
would lead to the strange result that every defendant
who had not turned himself in immediately upon com-
mission of his crime would receive the enhancement.
Whether a decision not to turn oneself in forthwith
after committing the crime is attributable to panic,
instinct, or calculation is neither here nor there. It is the
conduct, not the state of mind, that distinguishes
initial flight from obstructive conduct. True, the obstruc-
tion of justice guideline requires “willful” obstruction of
12 Nos. 08-3528, 09-1529, 09-1631
justice, U.S.S.G. § 3C1.1, and the cases (naturally) echo
this. E.g., United States v. Arceo, supra, 535 F.3d at 687;
United States v. Draves, supra, 103 F.3d at 1338; United States
v. Burton, supra, 933 F.2d at 918. But because of limitations
on mind reading, willfulness usually has to be inferred
from conduct rather than being determined directly. E.g.,
Ratzlaf v. United States, 510 U.S. 135, 149 n. 19 (1994)
(“willfulness ‘is usually established by drawing reasonable
inferences from the available facts,’ ” quoting United States
v. Bank of New England, N.A., 821 F.2d 844, 854 (1st Cir.
1987)); United States v. Scott, 660 F.2d 1145, 1160 (7th Cir.
1981); United States v. Beidler, 110 F.3d 1064, 1069 (4th Cir.
1997). If you shoot someone between the eyes at 100 yards
with a sniper rifle, your testimony that it was an accident
is unlikely to be believed. In like vein United States v.
Reeves, 586 F.3d 20, 26 (D.C. Cir. 2009), observed that
“because [the defendant’s] conduct was inherently ob-
structive, the [district] court was permitted to infer
Reeves’ willfulness, and therefore it would have been
futile to argue Reeves lacked the subjective intent to
obstruct justice based on the lack of a court order.”
Obstruction generally and in this case begins when
there has been no initial flight (remember that Hernandez
was cooperating with the police until he decided to
decamp for Florida), or when flight ends. But there is
“flight plus”—unusually elaborate, pertinacious, or
dangerous efforts to avoid being captured in the first
place, as in such cases as United States v. White, 903 F.2d
457, 461-63 (7th Cir. 1990); see also United States v. Bliss,
430 F.3d 640, 647-51 (2d Cir. 2005); United States v. Walcott,
supra, 61 F.3d at 639; United States v. Madera-Gallegos, supra,
Nos. 08-3528, 09-1529, 09-1631 13
945 F.2d at 267-68. Hernandez was released in exchange
for promises to cooperate and to keep in touch, broke
his promises, created delay and expense, and so merited
the enhancement. See United States v. Arceo, supra, 535
F.3d at 687; United States v. Porter, supra, 145 F.3d at 903-
04; United States v. Walcott, supra, 61 F.3d at 639; United
States v. Mondello, 927 F.2d 1463, 1466-67 and n. 4 (9th Cir.
1991).
Hernandez’s last argument is that he should not have
been denied a sentencing discount for acceptance of
responsibility just because he was found to have
obstructed justice. He is right that a finding of obstruc-
tion of justice does not automatically preclude a finding
that the defendant accepted responsibility for his crime.
In United States v. Buckley, 192 F.3d 708, 711 (7th Cir. 1999),
we gave the example of a defendant who when first
questioned by the police had denied possessing an
illegal weapon but the next day gave a complete con-
fession and later pleaded guilty. We said that he might
deserve an enhancement for the obstruction of justice
yet at the same time earn a discount for having fully, if
slightly belatedly, accepted responsibility for his crime.
And in United States v. Lallemand, 989 F.2d 936, 938 (7th
Cir. 1993), we upheld the enhancement over the defen-
dant’s objection that the award to him of the sentencing
discount for acceptance of responsibility should have
precluded a sentencing increase for obstruction of justice.
We said, as in Buckley, that one “can have the act of ob-
struction or attempt to obstruct at time t, and the accep-
tance of responsibility at time t + 1. There is no logical or
practical incompatibility, and no barrier in the language of
14 Nos. 08-3528, 09-1529, 09-1631
the guidelines.” See also United States v. Travis, 294 F.3d
837, 841 (7th Cir. 2002); United States v. Mayberry, 272
F.3d 945, 949 (7th Cir. 2001).
A note to the sentencing guidelines states that when
a defendant receives a sentencing enhancement for ob-
struction of justice the case must be “extraordinary” to
warrant a discount for acceptance of responsibility.
U.S.S.G. § 3E1.1, Application Note 4; see, e.g., United
States v. Keeter, 130 F.3d 297, 299 (7th Cir. 1997); United
States v. Jeross, 521 F.3d 562, 581-82 (6th Cir. 2008);
United States v. Campos, 362 F.3d 1013, 1016-17 (8th Cir.
2004). The judge in the present case said the case was
not extraordinary and the government says the case
was not extraordinary; neither offered any explanation
for so concluding but we suppose it seemed obvious to
them—and it is obvious. The purpose of the sentencing
discount for acceptance of responsibility is to reduce the
burdens of trial to prosecutors, judges, victims, jurors,
and witnesses by inducing defendants to plead guilty.
United States v. Woodard, 408 F.3d 396, 397-98 (7th Cir.
2005); United States v. Morgano, 39 F.3d 1358, 1377-78
(7th Cir. 1994); United States v. Tolson, 988 F.2d 1494, 1498-
99 (7th Cir. 1993). The distinct notion that judges can
determine from words rather than deeds (deeds such
as pleading guilty) whether a defendant is truly
remorseful is increasingly recognized to be implausible.
Michael M. O’Hear, “Appellate Review of Sentences:
Reconsidering Deference,” 51 Wm. & Mary L. Rev. 2123,
2142-47 (2010); O’Hear, “Remorse, Cooperation, and
‘Acceptance of Responsibility’: The Structure, Implementa-
tion, and Reform of Section 3E1.1 of the Federal Sentencing
Nos. 08-3528, 09-1529, 09-1631 15
Guidelines,” 91 Nw. U. L. Rev. 1507, 1554-56 (1997); Bryan
H. Ward, “Sentencing Without Remorse,” 38 Loyola U.
Chi. L.J. 131, 133-36, 164-67 (2006); Craig S. Lerner, “Rea-
sonable Suspicion and Mere Hunches,” 59 Vand. L. Rev. 407,
450-51 (2006). As one district judge (unidentified) is quoted
as saying in Stanton Wheeler, Kenneth Mann & Austin
Sarat, Sitting in Judgment: The Sentencing of White-Collar
Criminals 117 (1988), “If you give too much consideration
to it [remorse] then you are a sitting duck, I suppose, for
sham protestations of remorse and breast-beating, and
buckets of tears and appeals of sympathy.”
We thus noted in United States v. Beserra, 967 F.2d 254,
256 (7th Cir. 1992), our preference for “deeds over
words—external, verifiable, expiatory acts over self-
serving, unverifiable reports of interior mental states.
Not only are deeds better evidence than words (’putting
your money where your mouth is’), but they have value
to the law-enforcement authorities, compared to which
breast-beating before the sentencing judge is a debased
currency indeed. It is better to credit a defendant for
doing something of value to someone than for retaining
a lawyer who can help him craft a spiel that will tread
the delicate line between making excuses and confessing
a will to evil. Bessera has no expiatory deeds.” In a case
such as this in which a two-year manhunt is necessary
to bring the defendant to trial after he procured his
release from jail by promising to cooperate and keep in
touch, the burden on law enforcement imposed by the
obstruction of justice is bound to exceed any plausible
benefit to law enforcement from his plea of guilty. The
two-level reduction in Hernandez’s offense level that
16 Nos. 08-3528, 09-1529, 09-1631
he sought for acceptance of responsibility would if
granted have negated the two-level enhancement that he
received for obstruction of justice; his initial cooperation
would have bought him the right to become a fugitive
from justice.
A FFIRMED.
6-15-10