After a trial to the court, appellant was convicted of violating 13 U.S.C. § 221(a) ^[1] by refusing to answer questions on the Department of Commerce census form of 1970. The court sentenced him to pay a $50 fine. We reverse.

Steele raises several points, only two of which require discussion: (1) answering the census questions would have required him to incriminate himself, and (2) the authorities singled him out for prosecution because he had publicly advocated noncompliance with census requirements.

SELF-INCRIMINATION

Appellant and six other unrelated young adults resided in a private home in Honolulu. When a census enumerator called for his questionnaire, Steele refused to answer all questions on the form ^[2] and returned it to the enumerator with two pages removed. Next, a supervisor called and was similarly rebuffed. Finally, after a regional officer had tried and failed to interview Steele, the effort was abandoned. Steele testified at trial that he had refused to answer on constitutional grounds. ^[3]

Steele argues that he may have been in violation of the Honolulu Zoning Code because more than five unrelated people lived in his single-family dwelling. Answering the questionnaire would have disclosed this fact, and might have subjected him to criminal prosecution by municipal authorities.

The argument calls into question the impact of the use immunity provisions of the census statutes. It is provided in 13 U.S.C. § 8(c): “In no case shall information furnished under the authority of this section be used to the detriment of the persons to whom such information relates.” And 13 U.S.C. § 9(a) provides: “Copies of census reports which have been so retained shall be immune from legal process, and shall not, without the consent of the individual . concerned, be admitted as evidence or used for any purpose in any action, suit, or other judicial or administrative proceeding.”

The government argues that these provisions shield Steele from any substantial risk of self-incrimination. Steele disagrees. We need not resolve that conflict, because we have determined that Steele’s second point has merit and requires reversal of his conviction.

DISCRIMINATORY PROSECUTION

Steele claims the census authorities deliberately applied an unjustifiable standard in selecting offenders for prosecution under 13 U.S.C. § 221(a). Only four people in Hawaii were chosen for prosecution. All had participated in a census resistance movement, publicizing a dissident view of the census as an unconstitutional invasion of privacy and[*1151] urging the public to avoid compliance with census requirements.

Steele held a press conference, led a protest march, and distributed pamphlets entitled “Big Brother is Snooping.” David Watamull was the owner of radio station KTRG, which broadcast editorials on the census. Census authorities had complained to the Federal Communications Commission about them because they “were calculated to incite people to subvert the census law.” Donald Dickinson spoke against the census as an announcer on station KTRG. William Danks headed the state chapter of a group called Census Resistance ’70; he distributed pamphlets and publicly criticized the census.

Leland Gray, the Regional Technician for the census in Hawaii, described the four as “hard core resisters.” He ordered his staff to compile special background dossiers on them, a discretionary procedure not followed with any other offenders. Gray testified that his organization had been very concerned about the census resistance movement.

Steele attempted to prove that many others in Hawaii had provided census officials with no more information than he had. In a motion for a bill of particulars, he asked the government how many others in Hawaii had committed the same offense. The United States Attorney’s office replied that the information was not available. Steele himself located six other persons who had completely refused on principle to complete the census forms. None of those had taken a public stand against the census and none were recommended for prosecution.

Mr. Gray testified that, to the best of his recollection, the four men prosecuted were the only ones who completely refused to cooperate. Steele’s evidence about the six others demonstrates that Mr. Gray’s memory was faulty.

Yick Wo v. Hopkins, 118 U.S. 356, 6 S.Ct. 1064, 30 L.Ed. 220 (1886), established the principle that equal protection of the law is denied when state officials enforce a valid statute in a discriminatory fashion. ^[4] The Due Process Clause of the Fifth Amendment furnishes a federal defendant with the same guarantee against discriminatory federal prosecution. Washington v. United States, 130 U.S.App.D.C. 374, 401 F.2d 915 (1968); cf. Bolling v. Sharpe, 347 U.S. 497, 74 S.Ct. 693, 98 L.Ed. 884 (1954). A defendant cannot be convicted if he proves unconstitutional discrimination in the administration of a penal statute. Two Guys from Harrison-Allentown, Inc. v. McGinley, 366 U.S. 582, 588, 81 S.Ct. 1135, 6 L.Ed.2d 551 (1961).

Mere selectivity in prosecution creates no constitutional problem. Oyler v. Boles, 368 U.S. 448, 456, 82 S.Ct. 501, 7 L.Ed.2d 446 (1962). To invoke the defense successfully one must prove that the selection was deliberately based on an unjustifiable standard, such as race, religion, or other arbitrary classification. Oyler v. Boles, supra, 368 U.S. at 456, 82 S.Ct. 501. See Snowden v. Hughes, 321 U.S. 1, 64 S.Ct. 397, 88 L.Ed. 497 (1944); Rhinehart v. Rhay, 440 F.2d 718 (9th Cir. 1971); People v. Utica Daw’s Drug Co., 16 A.D.2d 12, 225 N.Y.S.2d 128 (1962). Steele is entitled to an acquittal if his evidence proved that the authorities purposefully discriminated against those who chose to exercise their First Amendment rights.

Although hampered by the government’s refusal to supply data on the number of like offenses, Steele did manage to show that at least six others had committed the same offense. The Re[*1152] gional Technician said he had never heard of any of the six. The trial judge concluded that Steele and his colleagues were the only violators known to the census officials. That finding is not supported by the record taken as a whole.

As Mr. Gray explained them, the census operating procedures would in the normal course of events furnish information about any person who failed to complete the questionnaire. A refusal would be reported in the chain of command, from the enumerator through a Crew Leader, a Field Supervisor, and a District Office Manager, to the Regional Technician, Mr. Gray. At least two officials would attempt to obtain the missing answers from the violator. The system would reveal the names of offenders, and visits by census officials would lay the factual foundation for proving specific criminal intent.

This information-gathering system should have apprised the Regional Technician of the names of all who refused to complete the questionnaire. Yet Mr. Gray recollected only four total refusals, while the evidence establishes a minimum of ten. That fact alone strongly suggests a questionable emphasis upon the census resisters. When one also considers that background reports were compiled only on persons who had publicly attacked the census, the inference of discriminatory selection becomes almost compelling. An enforcement procedure that focuses upon the vocal offender is inherently suspect, since it is vulnerable to the charge that those chosen for prosecution are being punished for their expression of ideas, a constitutionally protected right.

The government offered no explanation for its selection of defendants, other than prosecutorial discretion. That answer simply will not suffice in the circumstances of this case. Since Steele had presented evidence which created a strong inference of discriminatory prosecution, the government was required to explain it away, if possible, by showing the selection process actually rested upon some valid ground. Mere random selection would suffice, since the government is not obligated to prosecute all offenders, but no effort was made to justify these prosecutions as the result of random selection and Steele’s evidence was inconsistent with such a theory. Since no valid basis for the selection of defendants was ever presented, the only plausible explanation on this record is the one urged by Steele. We conclude that Steele demonstrated a purposeful discrimination by census authorities against those who had publicly expressed their opinions about the census.

The conviction is reversed.