United States v. Schipani

289 F. Supp. 43, 22 A.F.T.R.2d (RIA) 5416, 1968 U.S. Dist. LEXIS 12500

Opinion

OPINION AND ORDER

WEINSTEIN, District Judge.

The Supreme Court, upon being informed by the Solicitor General that the defendant, Joseph F. Schipani, was a participant in a number of conversations which had been electronically monitored by agents of the Federal Bureau of Investigation and which led to the use of tainted evidence against the defendant, vacated the defendant’s conviction for income tax evasion, and remanded “the cause to the district court for a new trial, should the Government seek to prosecute * * * anew.” 385 U.S. 372, 87 S.Ct. 533, 17 L.Ed.2d 428. Following a detailed analysis of the hundreds of exhibits and the testimony of the many witnesses relied upon at the trial, the government concluded that none of the evidence that it utilized was tainted and that “the Solicitor General was in ERROR” when he informed the Supreme Court to the contrary. Government’s Affidavit in Reply to Defendant’s Motion for Suppression of Evidence, p. 2 (Emphasis in original). Consequently, it has decided to prosecute anew.

Defendant has moved to suppress and a full evidentiary hearing has been held. The Government has freely revealed all data available to it. It has presented all the surveillance logs of conversations in which the defendant participated or was mentioned, the files of the Federal Bureau of Investigation, the Alcohol and Tobacco Tax Division of the Treasury Department, the Intelligence Division of the Internal Revenue Service, and the Department of Justice, and the testimony of the many agents and attorneys in the 1961-1963 investigations of the defendant and the 1966 review of their work by the Department of Justice. In addition, the Federal Bureau of Investigation “case agent” assigned to the Schipani investigation reviewed all the Federal Bureau of Investigation reports relating to the defendant and underlined all information obtained as a result of electronic surveillance.

This evidence has been examined to obtain answers to two questions. First, whether individual items of evidence used at the first trial were obtained, either directly or indirectly, as a result of illegal monitoring. Second, whether the entire investigation of the defendant was tainted because its intensity was substantially affected by the electronic surveillance. For the reasons stated below, the first question is answered in the affirmative and the second in the negative.

I. FACTS

A. Prior Proceedings

On October 15, 1965, after a non-jury trial, the defendant was found guilty in this Court of willfully attempting to evade the payment of income taxes for the years 1956 through 1960. 26 U.S.C. § 7201. The case was tried on a “net worth” theory. The evidence showed that the defendant had a certain net worth at the beginning of 1956 and an increase in net worth at the end of that year and each succeeding year during the indictment period. The defendant kept no books, failed to file returns for any of the years covered by the indictment, and adduced no proof at the trial.

The defendant’s financial practices were characterized by the trial court as being “marked by concealment through use of the names of others, failure to keep records in transactions when record keeping is usual and absence suspicious, extensive use of Manufacturers Trust Co. personal money orders, and large transactions in which currency was used.” United States v. Schipani, 63-CR-237, October 15, 1965, E.D.N.Y., p. 7. The most damaging evidence against the defendant were the money orders. Through them were traced many of defendant’s financial transactions and upon them the government based its computation of the defendant’s yearly gross income. In addition, the trial court found that the defendant had likely sources of income from an interest in a restaurant and as a negotiator in labor-management relations.

The conviction was affirmed by the Second Circuit (362 F.2d 825). Certiorari was denied on November 7, 1966 (385 U.S. 934, 87 S.Ct. 293, 17 L.Ed.2d 214).

On November 30, 1966, the Solicitor General filed a supplemental memorandum with the Supreme Court which suggested that “the Court vacate . the judgment . . . .” The basis for this recommendation was the discovery by attorneys in the Department of Justice that the defendant had been a participant in a number of conversations which had been electronically monitored as a result of a trespass by agents of the Federal Bureau of Investigation. This electronic surveillance began a short time prior to the commencement of the investigation of the defendant for tax violations and continued during the period when much of the evidence introduced at the defendant’s trial was uncovered.

In his memorandum to the Supreme Court, the Solicitor General stated:

“The substance of the monitored conversations involving petitioner and others was recorded in logs kept by the F.B.I. agents who conducted the electronic surveillance and was reflected in various F.B.I. reports which came to the attention of those in charge of investigating and prosecuting petitioner for possible tax violations. The reports did not disclose the manner in which the information contained therein had been obtained.

“Unlike the situation in Black v. United States, 385 U.S. 26, [87 S.Ct. 190, 17 L.Ed.2d 26] (1966), we cannot say in the instant case that none of the evidence used by the government at petitioner’s trial was obtained, either directly or indirectly, from an improper source. Some of the material in the F.B.I. reports (which were based in part on the electronic surveillance) provided investigatory leads and other information used in proceeding against petitioner. Since there was material evidence against petitioner which was tainted, his conviction cannot stand, and no purpose would be served here in having the district court conduct a collateral hearing, such as was suggested by the government in its memorandum in Black.”

In addition, the Solicitor General disclosed that while his memorandum was being prepared, it was learned that the Alcohol and Tobacco Tax Division had engaged in electronic surveillance of another establishment frequented by the defendant. On the information available to him at that time, which admittedly was incomplete, the Solicitor General advised the Court that it “appears that * * * no relevant information” was obtained from this source.

B. Roles of Government Agencies in Case

Four government agencies played a role in the events which culminated in the prosecution of the defendant for income tax evasion — the Federal Bureau of Investigation (F.B.I.), the Organized Crime and Racketeering Section of the Justice Department (Organized Crime Section), the Intelligence Division of the Internal Revenue Service (Intelligence), and the Alcohol and Tobacco Tax Division of the Treasury Department (Alcohol Tax Division). The role of each of them in the Schipani case will be examined with a view towards determining what effect, if any, the information obtained by the F.B.I. had on the investigations being conducted by the other three agencies.

1. F.B.I.

The F.B.I. was the first government agency to take an interest in the defendant. It had been compiling information on him since 1958. In an F.B.I. report dated October 10, 1960 (the last F.B.I. report compiled on the defendant prior to the start of the electronic surveillance), the defendant was described by one source as “one of the most influential and powerful figures in the New York underworld” and considered to be associated with several of New York’s top hoodlums. He was reported to have “inherited” the gambling interests of Joe Adonis and was suspected of being the owner of the Casa Bianca Restaurant and of having an interest in two others, the Hampshire House and the Gurney’s Inn. An earlier F.B.I. report identified him as being in control of an association of private carting companies. During the period between 1958 and 1960, the F.B.I. appears to have pursued its investigation of the defendant without any intention of initiating a criminal prosecution.

In January of 1961 the F.B.I. acquired a new and productive source of information concerning the defendant’s activities. As part of its investigation of one Michael Clemente, an associate of the defendant, the F.B.I. installed a microphone in the Prisco Travel Bureau, a “front” for Clemente’s criminal activities and a meeting place for many of New York’s top mobsters. This device was in operation from January 10, 1961 to January 31, 1963.

Although the defendant was not the subject of this electronic surveillance, nine conversations in which he was a participant and twenty-seven others in which he was discussed or his name was mentioned, were electronically monitored. The defendant’s voice was first picked up on March 2, 1961; on three occasions prior to that date, beginning on January 12,1961, his name was mentioned and his activities were discussed by Clemente and others.

The information obtained from this method of surveillance confirmed the F. B.I.’s belief that the defendant was intimately associated with many of the top figures in the Brooklyn underworld and revealed several of his current illicit activities. Numerous conversations were monitored concerning the defendant’s role in a private association of New York carting companies and his concealed interest in one such concern. In addition, information was obtained concerning the defendant’s involvement with the sale of Puerto Rican sweepstakes tickets on the New York docks, his partial ownership of the Casa Bianca Restaurant, and his control of a New York labor union local.

The further activities of the F.B.I. in the Schipani case were not developed. In February, 1962, it announced to the Alcohol Tax Division its intention to launch a saturation investigation of the defendant’s interests in the carting field. No evidence was adduced at the suppression hearing to indicate whether or not this plan was carried out, and, if so, with what results. No evidence, if any, resulting from the carting investigation by the F.B.I. was introduced at the trial.

2. Organized Crime and Racketeering Section of the Department of Justice

The appointment of Robert F. Kennedy as Attorney General in January of 1961 inaugurated a new phase in the fight against organized crime in the United States. The Attorney General decided that organized crime presented a national as well as a local problem and that only the federal government, with an organization as broad as that of the criminal syndicates, could effectively combat it.

To implement this concerted drive, an area coordinator in the Organized Crime Section was designated to select as “targets” for investigation those individuals suspected of being part of the organized crime network in the United States. The numerous federal law enforcement agencies involved in these investigations were ordered to cooperate and exchange information with each other and to synchronize their activities as much as possible. In order to facilitate this effort, all agencies were required to send in periodic progress and intelligence reports to the Organized Crime Section.

One of the targets selected for investigation as part of this program by the area coordinator, Philip Wilens, was the defendant. In a memorandum dated January 31, 1961, three weeks after the electronic surveillance had begun, Wilens recommended that two investigations of the defendant be initiated: one by Intelligence into the defendant’s “gambling activities in order to determine if there is available evidence of wagering tax violations” and the other by Alcohol Tax “to determine if Schipani has filed or caused to be filed falsified Forms 11 in applying for the retail liquor dealer stamps for these premises [Casa Bianca Restaurant, Hampshire House, and Gurney’s Inn].”

Wilens testified categorically that this decision was based solely upon information obtained from the untainted F.B.I. report of October 10, 1960. The data about the defendant’s suspected criminal activities in the Wilens memorandum was essentially a summary of this F.B.I. report. In addition, Wilens stated that this report constituted the only information he (and, to the best of his knowledge, any one else in his department) had about the defendant at that time and that he had no knowledge of any electronic surveillance of the defendant or any one connected with the case.

Wilen’s recommendations were accepted. By letters dated February 14, 1961 and February 27, 1961, the Alcohol Tax Division and Intelligence were requested to conduct investigations of the defendant. These letters were dictated by Wilens for his Chief’s signature sometime between January 31, 1961 and February 6, 1961. They are set out below (Silberling succeeded Hundley as Chief of the Organized Crime Section between the time the two letters were sent):

“2/14/61

Mr. Dwight E. Avis

Director, Alcohol and Tobacco

Tax Division

Internal Revenue Service

Washington 25, D.C.

Re: Joseph Frank Schipani, Brooklyn,

New York FBI #571946

Dear Mr. Avis:

This person is believed to have acquired undisclosed interests in the following retail liquor dealer establishments: Casa Bianca Restaurant and Cocktail Lounge, 4th Avenue and 100th Street, Brooklyn, New York and Hampshire House Restaurant, 339 Merrick Road, Rockville Center, Long Island, New York. His wife is believed to own 10 shares of stock in Guiney’s [sic] Inn, Inc., On the Brink of the Beach, Montauk Point, Long Island, New York.

The Attorney General has directed that we request your division to undertake investigations regarding whether Schipani has filed, or caused to be filed, falsified Forms 11.

In the event your files presently contain information relating to Schipani and his activities, we would appreciate receipt thereof. Also, if you desire further details concerning Schipani’s activities, they can be obtained by you from the Federal Bureau of Investigation.

Sincerely,

WILLIAM G. HUNDLEY

Special Assistant to the

Attorney General”

“February 27, 1961

Mr. H. Alan Long

Director

Intelligence Division

Washington, D. C.

Re: Joseph Frank Schipani,

Brooklyn, New York

FBI #571946

Dear Mr. Long:

The subject of this letter, Joseph Frank Schipani, is reputed to be a major racketeer in the Brooklyn, New York area and is said to have succeeded to the Brooklyn gambling interests of Joe Adonis.

In this regard he is supposed to be operating a handbook in the area of 42nd Street and 3rd Avenue, Brooklyn, New York, which covers the entire Bay Ridge section. He is also said to have an interest in a handbook operating in the general area of 5th Avenue in Brooklyn.

The Attorney General has directed that we request your Division to undertake an investigation of the activities of this person with regard to possible wagering tax violations.

If your files presently contain any information regarding this subject, we would greatly appreciate receipt thereof.

EDWIN SILBERLING

Chief, Organized Crime and

Racketeering Section

Criminal Division”

3. Investigations Conducted, by Alcohol and Tobacco Tax Division and Intelligence Division of Internal Revenue Service

(a) Liaison with F.B.I.

After the Supreme Court’s remand of this case, the Special Agent of Intelligence in charge of the Schipani investigation, Albert Masetti, together with a member of the United States Attorney’s staff, made a painstaking analysis of all the evidence introduced at the Schipani trial. The product of their labors was a comprehensive chart which denotes a specific source or sources for each witness and piece of evidence (including 292 exhibits) introduced at the trial.

This chart indicates that none of the agents involved in the investigation obtained leads from the electronic surveillance by the F.B.I. and that all the leads they did get were through legitimate techniques — routine investigatory procedure, a mail watch, and physical surveillance and observation of the defendant. Special Agent Masetti testified that he received no leads other than those indicated on the chart. The defendant has stipulated that the chart is accurate insofar as the leads shown on it were used, but not insofar as it suggested that there were no other leads.

Masetti and the Alcohol Tax Division agents assigned to the case emphatically maintained that they had no knowledge, either personal or hearsay, of any electronic surveillance being conducted by the F.B.I., although they had met with F.B.I. officials in New York several times during the course of their investigations. Masetti’s predecessor as agent-in-charge of the Schipani investigation reported in his work sheet that on March 22, 1961 he examined the files of the F.B.I.; however, a reading of his notes indicates that he was only shown the untainted F.B.I. report of October 10, 1960.

The several F.B.I. agents who had been involved in the Schipani case corroborated this testimony; they stated that they had never informed Intelligence or the Alcohol Tax Division of any electronic surveillance. Further, and more to the point, they testified categorically that, in spite of the Attorney General’s directive that there be a full exchange of information and cooperation between government law enforcement agencies, it was the practice of the F.B.I. at that time not to disclose to any other agency any information obtained through electronic surveillance. As the supervisor of the F.B.I. criminal intelligence squad in New York, James P. Mansfield, put it, “It was our policy at that time not to disseminate information obtained from technical sources.” Such information, it was explained, was only available within the F.B.I. itself on a “need-to-know” basis.

Despite this independent policy of the F.B.I., it is not improbable that the information obtained through electronic surveillance was funnelled up to the Organized Crime Section in Washington and then down to local agents of Intelligence and the Alcohol Tax Division. As the Solicitor General pointed out in his brief, such information was reflected in the periodic reports prepared by the F. B.I. on the defendant. Copies of these reports were sent to the F.B.I.’s Washington office. The F.B.I. agents who testified at the hearing could not state that there was no dissemination by the Washington office. Intelligence reports were regularly filed by the F.B.I. with the Organized Crime Section as part of the Attorney General’s drive on organized crime and it is possible that these reports, if not the same ones sent to F.B.I. headquarters by its New York office, nevertheless contained information obtained through the electronic surveillance of the Prisco Travel Bureau.

This conclusion is supported by an examination of the Alcohol Tax Division status reports on the Schipani investigation, all of which were forwarded to Special Agent Masetti and the Attorney General’s office. These reports are replete with references to information received from a “confidential source” — possibly the F.B.I.’s electronic device — concerning the defendant’s involvement in the carting industry. Also contained in these reports are descriptions of several conferences with Gerald Shur, the successor to Philip Wilens as Area Coordinator, and instances of intelligence data being received from the Organized Crime Section in Washington. For example, the report dated June 23, 1961, states that “[d]uring the past period a communication has been received * * * from the Chief, Organized Crime * * * Section * * * to the effect that Benny Ladner, Ralph Russo, Ralph “The Chef,” and Joseph Schipani are partners in the Casa Bianca.” And, in the report of March 1, 1962, mention is made of a conference held in the office of the Assistant Regional Commissioner of the Alcohol Tax Division on February 14, 1962, at which “it was decided that the investigation into the carting companies be suspended for approximately six weeks to prevent conflicts with the F.B.I.’s plan to launch a saturation investigation of Joseph Schipani’s interest in this field.”

The attorney in the Department of Justice who conducted the 1966 review of the Schipani case, Dewey O’Brien, reached the conclusion that there was substantial cooperation among the several agencies involved in the case. In his memorandum dated November 18, 1966, he noted, for example, that on July 12, 1961 — five days before the I.R.S. was requested to conduct a “saturation” investigation of the defendant “because of information ‘supplied us by other investigative agencies’ ” — Shur reported “on a field trip on the case in which he talked to agents of the different agencies.” O’Brien wrote:

“In a letter of June 9, 1961, from Silberling to Avis of ATU, Silberling said it had come to their attention that Benny Ladner, listed as a witness in SAR (and others) and Schipani were partners in Casa Bianca; that ‘it has been reported that Michelino Clemente has been attempting to purchase Casa Bianca over a period of several months and has been dealing with Schipani.’ In a memo, Shur reported (7/12/61) on a field trip on the case in which he talked to agents of the different agencies.

By letter of July 17, 1961, Mr. Miller asked Mr. Caplin for a ‘saturation investigation’ because of information ‘supplied us by other investigative agencies.’ Another letter (10/19/61) to Alan Long gave a full report on Schipani and said ‘the following information was obtained from recent reports of the FBI’ and referred to the Casa Bianca and the report that Schipani controlled the garbage disposal unions in Brooklyn.’.’

A finding that information obtained from electronic surveillance may have been transmitted to Intelligence and the Alcohol Tax Division does not reflect doubt about the credibility of Philip Wilens or the various agents who testified that they were unaware of such surveillance. The source or sources for data contained in F.B.I. reports is denoted by codes which are not appended to a report when it is circulated outside the agency; only persons within the F.B.I. could determine the source of intelligence information contained in these reports. Thus, Wilens, upon being shown an F.B.I. report based in part upon electronic surveillance, testified that he would not be able to tell whether any particular entry in the report was derived from this-method of surveillance and that he might make such information available to another agency without being aware of its source. In turn, Intelligence and Alcohol Tax agents admitted that they had received leads from Washington without knowing their source.

(b) Alcohol and Tobacco Tax Division Investigation

The Alcohol Tax Division played a decisive role in amassing the evidence uti lized at the defendant’s trial, particularly with regard to the crucial money order lead. As noted above, on February 14, 1961, Dwight E. Avis, the Director of the Alcohol Tax Division, was requested by the Organized Crime Section to begin an investigation into the defendant’s suspected undisclosed interests in three restaurants. On March 3, 1961, Avis forwarded the letter to the Alcohol Tax Division’s Assistant Regional Commission in New York with instructions to conduct a “top priority” investigation, but within narrowly confined limits:

“Mar — 3 1961

Assistant Regional Commissioner

Alcohol and Tobacco Tax

New York, New York

As a part of the Government’s drive against organized crime the Attorney General has requested this Division to give top priority to the investigation of the possible undisclosed interests which certain major racketeers may have in the ownership and operation of liquor establishments.

******

* * * We also believe that your investigation should be confined to the development of such evidence or information as may be obtained by inspection of the premises, examination of the records in your office, other segments of the Service, the State or local licensing authorities, and public utilities, and interviews with the landlord, informers, and wholesale liquor dealers or any others having business dealing with the establishments, etc. In short, any inquiry short of an audit or other detailed examination of the financial books and records of the establishments and individuals as it is believed that such aspects can better and more properly be made by the Intelligence and Audit Divisions.

******

The investigation and prompt completion of this or any other work in connection with the Attorney General’s drive against organized crime should be given special attention and priority over all other assigned duties even though it may adversely affect your regular enforcement operations. * * *

Dwight E. Avis

Director, Alcohol and Tobacco Tax Division”

The Alcohol Tax Division was supplied with leads to the Casa Bianca, the Hampshire House, and Gurney’s Inn and presumably given access to the F.B.I.’s untainted October 10, 1960 report on the defendant.

After a preliminary inquiry into the leads and a conference with Gerald Shur in Washington, the Alcohol Tax Division decided to broaden its investigation and place the defendant under constant surveillance. This surveillance began on May 17, 1961, and at times six to eight men were assigned to tail the defendant. It produced immediate results. The defendant was observed frequenting the Casa Bianca, driving a car registered in another person’s name, enjoying a relatively high standard of living, paying bills in cash, and most important, on July 17, 1961 and again on September 25, 1961, purchasing money orders at the Manufacturers Trust Co. bank. By October 27, 1961, a list of money orders that the defendant had purchased from Manufacturers Trust had been compiled.

Meanwhile, a routine search disclosed that the defendant had never filed an income tax return and, on June 1, 1961, Intelligence was asked to pursue this lead.

The defendant was also observed frequenting a health club located in a Brooklyn hotel. From August 21, 1961 to September 8, 1961, the Alcohol Tax Division engaged in electronic surveillance of this establishment. The defendant’s voice was never picked up and no information was obtained pertaining to him.

(e) Intelligence Division of Internal Revenue Service Investigation

On February 27, 1961, H. Alan Long, Director of Intelligence, was requested to conduct an investigation of the defendant’s gambling activities with regard to possible wagering tax violations. The case was assigned to Intelligence’s Brooklyn office in early March.

At the outset, there was considerable uncertainty as to the type and scope of the investigation to be conducted. Set out below is an excerpt from a memorandum, dated March 24, 1961, to the Assistant Regional Commissioner of Intelligence of the New York City Region from the Chief of Intelligence in Brooklyn:

“Is it the desire of the National Office that we initiate wagering tax or income tax investigations, or a combined investigation involving both areas of tax?

In view of the nature of the request (Department of Justice) shall we initiate a saturation type investigation including surveillance, etc., or should the investigations be restricted in time and manpower, pending more explicit instructions ?”

One day earlier, Intelligence’s Assistant Regional Commissioner had been notified by H. Alan Long that, at the present time, a saturation investigation was not contemplated. This information was immediately conveyed to the Brooklyn office, along with the instruction that “in view of * * * [the defendant’s] alleged illegal activity in gambling it is presently assumed that * * * [an investigation] to determine if * * * [he has] violated the wagering tax provisions is more appropriate.”

Both of these decisions concerning the nature of the investigation — that it be non-saturation and focus on the defendant’s gambling activities — were soon to be changed. On July 17, 1961, the I.R.S. was requested by the Organized Crime Section to intensify its investigation of the defendant and to classify it as a saturation investigation:

“Honorable Mortimer M. Caplin

Commissioner

Internal Revenue Service

Washington 25, D. C.

Dear Mr. Caplin:

Based upon discussions with the Intelligence Agents assigned to the [case of] * * * Joseph Frank Sehipani and further based upon information supplied to us by other investigative agencies,‘it appears that the Internal Revenue investigation * * * will be productive.

It is therefore respectfully requested that the investigations of the above subjects be classified as saturation investigations.

Sincerely,

HERBERT J. MILLER, JR.

Assistant Attorney General”

There is no indication in the record as to what information was supplied by “other investigative agencies.”

Meanwhile the Intelligence probe had already undergone a major change — its transformation from a gambling tax to an income tax investigation. The Sehipani investigation had been transferred to Intelligence’s Manhattan district office on April 28, 1961 and assigned to Special Agent Masetti. Masetti was “unfamiliar” with wagering tax cases; income tax fraud was his field. Consequently, although he was aware of the directive to proceed along other lines, he decided to conduct an income tax investigation. Masetti testified that this decision was made solely by him, without his consulting anyone.

Masetti proceeded to conduct a long, arduous and productive investigation into the defendant’s financial affairs. Certain aspects of this inquiry have already been mentioned. No useful purpose would be served by further relating its details.

C. Review by Department of Justice in 1966

The fact that electronic surveillance had been employed in the Sehipani investigation did not become known to attorneys in the Department of Justice until late October, 1966. The Supreme Court, without yet being aware of this information, denied certiorari on November 7, 1966, and the defendant was slated to begin serving his sentence shortly. An immediate review of the case was undertaken by Dewey O’Brien, then the Assistant Chief of the Criminal Section of the Tax Division. On November 18, 1966, O’Brien completed his review and sent a six-page memorandum containing his findings and conclusion to Mitchell Rogovin, Assistant Attorney General in charge of the Tax Division.

In his memorandum, O’Brien stated that he “would be reluctant to recommend outright dismissal of the indictment at this point without a more painstaking examination of the evidence” since many leads were available before the electronic surveillance began, but he could not “see how we can support the conviction.” The basis for the latter conclusion was his findings that the electronic surveillance, which had been contemporaneous with the tax evasion investigation, had revealed likely sources of defendant’s income and that there had been exchanges of information between the various agencies involved. As Mr. O’Brien ironically pointed out:

“during the period in which the Schipani tax case was investigated the Government achieved the rare objective of complete interchange of information between three investigative agencies and the Department of Justice. Unfortunately, much of it was tainted.”

O’Brien’s opinion was based upon an examination of the F.B.I. reports on the defendant, the surveillance logs, the files of Intelligence, the Alcohol Tax Division and the Organized Crime Section, and the trial court’s findings of fact and conclusions of law (but not the trial transcript). In addition, he had discussed the case with several of the Intelligence and Alcohol Tax Division agents involved. He did not speak to any of the F.B.I. agents and, thus, he may have been unaware of the limited cooperation provided by the F.B.I.’s New York office.

O’Brien did not write the supplemental brief which the Solicitor General submitted to the Supreme Court on November 30,1966. He acknowledged that the position taken by the Solicitor General was somewhat at variance with his own, but he was at a loss to account for it. His then position was best summed up, he testified, by the Solicitor General’s statement that “we cannot say in the instant case that none of the evidence used by the government at petitioner’s trial was obtained, either directly or indirectly, from an improper source.”

Asked by the Court to explain the Solicitor General’s statement that “Some of the material in the F.B.I. reports (which were based in part on the electronic surveillance) provided investigatory leads and other information used in the proceeding against petitioner” and to identify the material and leads to which the Solicitor General referred, O’Brien stated :

“I do not think I would have phrased it that way if I would have written it.

I would have said, ‘May have provided’ because at that time we could not tell exactly what was done with the information.

There was for example, an earlier report of the FBI before this — before the intercepted information — which had some of the same information as contained in one of the intercepts, and * * * the question would have been, whether or not the information in the tax case stemmed from the intercept or from that earlier report.

I did not have the answer.”

In addition, O’Brien pointed out that since he had not read the transcript of the defendant’s trial, he could not say whether any specific item of evidence was tainted.

O’Brien assumed that the Solicitor General’s brief had at least been partially based upon his memorandum. He did not know if the Solicitor General had additional information or had conducted an independent review. There is no indication in the record that such a review had taken place.

In order to assess the significance of this evidence, we turn now to the principles of law by which evidence must be evaluated on a motion to suppress. The key issues are those of burdens of proof.

II. APPLICABLE LAW

A. Particular Items of Evidence

1. Is evidence to be excluded if it is obtainted as a combined result of tainted and untainted leads?

The test laid down in Silverthorne is that evidence is admitted if it is obtained from an untainted “independent source” but excluded if it is obtained as the result of illegality. Silverthorne Lumber Co. v. United States, 251 U.S. 385, 392, 40 S.Ct. 182, 64 L.Ed. 319 (1920). In the case before us a more appropriate simile than the “fruit of the poisonous tree” (Nardone v. United States, 308 U.S. 338, 341, 60 S.Ct. 266, 84 L.Ed. 307 (1939)), is, perhaps, that of a tree nourished by both pure and polluted waters. In this circuit the tendency has been to find any appreciable pollution sufficient to taint the fruit. See United States v. Paroutian, 299 F.2d 486 (2d Cir. 1962).

If the evidence is discovered as a result of both legal and illegal leads, it is inadmissible. It does not matter that the legal lead would itself probably have sufficed to uncover the evidence. As the Court of Appeals for this circuit declared, it is not sufficient for the government to show that “in the normal course of events it might have discovered the questioned evidence without an illegal search”:

“a showing that the government had sufficient independent information available so that in the normal course of events it might have discovered the questioned evidence without an illegal search cannot excuse the illegality or cure tainted matter. Such a rule would relax the protection of the right of privacy in the very cases in which, by the government’s own admission, there is no reason for an unlawful search. The better the government’s case against an individual, the freer it would be to invade his privacy. We cannot accept such a result. The test must be one of actualities, not possibilities.” United States v. Paroutian, 299 F.2d 486, 489 (2d Cir. 1962).

See also Pitler, “The Fruit of the Poisonous Tree” Revisited and Shepardized, 56 Calif.L.Rev. 579, 624-30 (1968) (“The logic of the ‘inevitable discovery’ . . . collides with the fundamental purpose of the exclusionary rule”); Note, Fruit of the Poisonous Tree Doctrine — A plea for Relevant Criteria, 115 U. of Pa.L.Rev. 1136, 1143 (1967) (defendant need not show that police “would not have discovered the challenged evidence ‘but for’ the illegal police conduct”).

2. What is the burden of proof?

(a) Common Law Rule

Once it becomes apparent that the issue is in the case (Maguire, Evidence of Guilt, 221 (1959)), the government bears the burden of convincing the court that evidence it seeks to introduce at a criminal trial was not obtained by it in violation of a defendant’s constitutional rights.

There is lack of agreement about the weight of this burden. Some have suggested that it is by a preponderance. Clifton v. United States, 125 U.S.App.D.C. 257, 371 F.2d 354, 358-359 (1966), cert. denied, 386 U.S. 995, 87 S.Ct. 1312, 18 L.Ed.2d 341 (1967) (confession). Others use a variant of the phrase “clear and convincing evidence.” United States v. Smith, 308 F.2d 657, 663 (2d Cir. 1962), cert. denied, 372 U.S. 906, 83 S.Ct. 717, 9 L.Ed.2d 716 (1963) (consent to search must be “proven by clear and positive evidence”); Fisher v. United States, 382 F.2d 31, 34 (5th Cir. 1967) (confession; “clear cut determination”); Evans v. United States, 375 F.2d 355, 360 (8th Cir. 1967), rev’d on other grounds sub nom. Bruton v. United States, 391 U.S. 123, 88 S.Ct. 1620, 20 L.Ed.2d 476 (1968) (confession); United States v. Page, 302 F.2d 81, 83-84 (9th Cir. 1962) (consent to search; “clear and positive testimony”); cf. United States v. Wade, 388 U.S. 218, 240, 87 S.Ct. 1926, 1939, 18 L.Ed.2d 1149 (1967) (government could “establish by clear and convincing evidence that the in-eourt identifications were based upon observations of the suspect” other than those made at the prior identification; reasons for standard not discussed). Still others have declared that “proof beyond a reasonable doubt” is required. Mullins v. United States, 382 F.2d 258, 261 (4th Cir. 1967) (confession); United States v. Inman, 352 F.2d 954, 956 (4th Cir. 1965) (confession); Clifton v. United States, 125 U.S.App.D.C. 257, 371 F.2d 354, 360-364 (1966), cert. denied, 386 U.S. 995, 87 S.Ct. 1312, 18 L.Ed.2d 341 (1967) (concurring opinion) (confession) ; Pea v. United States, 397 F.2d 627, 633, n. 7 (D.C.Cir.1968), (collecting cases in other jurisdictions; confession) ; United States v. Taglianetti, 274 F.Supp. 220, 226 (D.R.I.1967) (electronic surveillance). Cf. Gillison v. United States, 399 F.2d 586, 588 (D.C.Cir. July 2, 1968) (state must show that line of questioning was harmless beyond reasonable doubt); United States v. Feinberg, 383 F.2d 60, 70, n. 10 (2d Cir. 1967), cert. denied, 389 U.S. 1044, 88 S.Ct. 788, 19 L.Ed.2d 836 (1968) (citing New York standard of beyond a reasonable doubt in confession cases) ; People v. Johnson, 68 Cal.Rptr. 599, 609, 441 P.2d 111, 121 (May 28, 1968) (state must show beyond reasonable doubt that erroneous admission of prior inconsistent statement constituted harmless error). Mathes and Devitt, Federal Jury Practice and Instructions, § 8.16 (1965) (jury to be instructed not to use confession unless satisfied beyond a reasonable doubt that it was voluntary). Finally, there are opinions eschewing the opportunity to fix a standard. See, e. g., Jackson v. Denno, 378 U.S. 368, 405, 84 S.Ct. 1774, 12 L.Ed.2d 908 (1964) (dissent) (question of weight of burden in confession cases left open); Pea v. United States, 397 F.2d 627, 639 (D.C.Cir.1968) (dissent) (Supreme Court’s refusal “to come to grips with problem”; others “have avoided the problem”); cf. United States v. Jordan, 399 F.2d 610, 614 (2d Cir. July 30, 1968) (“government had a heavy burden of proving voluntary consent” to search); Carlo v. United States, 286 F.2d 841, 848 (2d Cir.), cert. denied, 366 U.S. 944, 81 S.Ct. 1672, 6 L.Ed.2d 855 (1961) (search and seizure; “quantum of proof * * * not that required on a trial of the issue of guilt or innocence”) .

In the case before us, the precise weight of the government’s burden may affect the decision on the motion to suppress. We are, therefore, compelled to address the question with some attention. The matter presents inherent practical and theoretical difficulties.

The problem will, perhaps, be more clearly exposed by considering a number of simple hypothetical situations. (The terminology used generally follows Michael and Adler, The Trial of an Issue of Fact, 34 Colum.L.Rev. 1224, 1252 (1934) and Michael and Adler, The Nature of Judicial Proof (1931). See also, e. g., Ball, The Moment of Truth: Probability Theory and Standards of Proof, 14 Vand.L.Rev. 807 (1961); James, Relevancy, Probability and the Law, 29 Calif.L.Rev. 689 (1941)).

In a civil case involving a breach of contract where defendant claims that he did not execute the agreement, the court will admit the alleged written contract if a reasonable man might find that it was signed by or on behalf of the defendant. Admissibility depends purely on the document’s meeting minimum standards of probative force. There may be mixed questions of law and fact relating to “execution” turning on whether, for example, the document was forged or whether the defendant was tricked into signing it. In determining whether the material proposition — the defendant “entered” into the contract— has been established, the jury will consider all of the admitted evidence, including the document before it. If there is some doubt about whether the defendant’s signature was forged, or whether he had authorized an agent to execute it, the jury will consider this doubt in evaluating all the lines of proof bearing on the material proposition.

The level of burden of proof (persuasion) with respect to the material proposition will be, since this is an ordinary civil case, a preponderance — that is to say the trier must be convinced, on the basis of his evaluation of the evidence, that the proposition is more probably true than false (50+% probable for purposes of this analysis). Thus, if the trier would have been just convinced that the material proposition was probably true absent this doubt about the authenticity of the document, doubt about this line of proof might sufficiently reduce his evaluation of the probability of the material proposition so that he might find that it was not proven (50% or less probable).

The jury’s evaluation of the evidence relevant to a material proposition requires a gestalt or synthesis which seldom needs to be analyzed precisely. Any item of evidence must be interpreted in the context of all the evidence introduced (and often of that reasonably expected which was not produced). In giving appropriate, if sometimes unreflective, weight to a specific piece of evidence the trier will fit it into a shifting mosaic. If the trier concludes that one item of evidence is reliable, this may affect his evaluation of the credibility of a witness who gives testimony inconsistent with this evidence. His evaluation of the credibility of this witness may, in turn, affect his conclusion as to the probative force of the witness’ testimony with respect to another line of proof. But confirming evidence of that other line of proof may require a reevaluation of the witness’ credibility and a complex readjustment of the assessment of all the interlocking evidence.

Since so much of the evaluation of evidence depends upon varying hypotheses applied by triers with different backgrounds and views of life, fact finding differences among jurors and between judge and jury are to be expected. The court’s function is, in the usual simple case, only to decide whether a reasonable man might have his assessment of the probabilities of a material proposition changed by the piece of evidence sought to be admitted. If it may affect that evaluation it is relevant and, subject to certain other rules, admissible. See Trautman, Logical or Legal Relevancy — A Conflict of Theory, 5 Vand.L. Rev. 385, 390 (1951). Even, therefore, if a juror decides that the probability is only 40% that the document referred to above is authentic, it may help him determine whether the material proposition is more probably true than not. This difference between the probabilities required of a material proposition and of relevant evidence and individual lines of proof partly explains why, in criminal eases, some courts have said that while the burden of proof of an element of the crime is beyond a reasonable doubt, individual items of evidence need not be proved by so high a standard. United States v. Valenti, 143 F.2d 362, 364 (2d Cir. 1943); People v. Klinkenberg, 90 Cal.App.2d 608, 204 P.2d 47, 62 (1948); 9 Wigmore, Evidence, § 2497, p. 324 (3d ed. 1940).

It is apparent, then, that the judge’s function in determining admissibility in a case such as is hypothesized above is relatively simple. 9 Wigmore, Evidence § 2550 (3d ed. 1940); Maguire and Epstein, Preliminary Questions of Fact in Determining the Admissibility of Evidence, 40 Harv.L.Rev. 392 (1927); Morgan, Some Problems of Proof Under the Anglo-American System of Litigation, 86-105 (1956); Note, 59 Harv.L.Rev. 488-491 (1964); California Law Revision Commission, Recommendation Proposing an Evidence Code, 56-70 (1965). He need not be concerned unduly about mistakenly admitting evidence because, if the document lacks probative force, the jury can (absent factors such as prejudice) be counted on to ignore it. Only time has been lost by admitting evidence with probative force so low that it cannot affect the outcome.

Where admissibility of evidence depends upon extrinsic policies rather than upon probative force, the problem is more complex and the pressure on the trial judge is greater. Let us suppose, solely for the sake of analysis, that the burden on the government is to prove by a preponderance that a proferred document with high probative force was not obtained in violation of the law. Assume that exclusion depends upon an ex trinsic policy designed to protect constitutional rights and to discourage obtaining evidence illegally. Suppose that the only material proposition in the case is whether defendant had certain knowledge (as that the powder he possessed was a narcotic). Let us suppose that the only evidence on the point is a letter from the defendant and that it would permit a jury to find beyond a reasonable doubt (95+% probable, for the purposes of this analysis) that he did have knowledge.

Now, if we suppose that the judge has decided that there is a question about whether the document was illegally obtained but that there is a bare preponderance (50+% probable) in favor of the government’s contention that it was legally obtained, the judge will, on the assumptions made, admit it. The question of whether the document was or was not illegally obtained is not considered by the jury and normally it will not have before it evidence bearing on this question. Thus, the jury will give the document its full probative weight and it will find the defendant guilty beyond a reasonable doubt.

Whether such a result in a criminal case is proper depends upon how the rule of law is stated. If it is: “the material proposition of knowledge must be proven beyond a reasonable doubt by evidence properly admitted in the trial” tne verdict should stand. The trier might have concluded that the probability of knowledge was 95+% if he assessed the document solely on the basis of its probative force. However, if it is: “the material proposition of knowledge must be proven beyond a reasonable doubt by legally obtained evidence,” there is doubt about the validity of the conviction. If the probative force is discounted by the probabilities relating to the illegality, then the total probative force of all the relevant evidence might well have fallen below 95+%. The documentary line of proof would not then have a force of 95+% probability but of (95+ x 50+) %, or somewhere in the order of 50+%. order of 50+%. See Ball, The Trial Court: Probability Theory and Jury Issues, in Communication Sciences and Law: Reflections from the Jurimetries Conference, 189-90 (Allen and Caldwell, eds., 1965).

This attenuation of probabilities results from the fact that, where variables are independent and the probability of a combination of variables is to be determined, the probability of each variable is multiplied by the probability of each other variable. See, e. g., Hoel, Elementary Statistics, 56-63 (1966); Keynes, A Treatise on Probability, 121, 135-136 (1952). Probative force and illegality are sufficiently independent to justify multiplying them for the purposes of this analysis. Where the probability of one of two variables approaches 100% the product will, of course, approximate the probability of the less probable variable.

Dicta of the Supreme Court, while not decisive, suggest that the rule of law is that a defendant must be proved guilty beyond a reasonable doubt by legally obtained evidence. See, e. g., Lynumn v. State of Illinois, 372 U.S. 528, 537, 83 S.Ct. 917, 9 L.Ed.2d 922 (1963); Payne v. State of Arkansas, 356 U.S. 560, 568, 78 S.Ct. 844, 2 L.Ed.2d 975 (1958); cf., Mesarosh v. United States, 352 U.S. 1, 9, 77 S.Ct. 1, 1 L.Ed.2d 1 (1956). For example, in Payne v. State of Arkansas, 356 U.S. 560, 568, 78 S.Ct. 844, 850, 2 L.Ed.2d 975 (1958), the Court noted that it:

“has uniformly held that even though there may have been sufficient evidence, apart from the coerced confession, to support a judgment of conviction, the admission in evidence * * * of the coreced confession vitiates the judgment * * *”

Cf. Fahy v. State of Connecticut, 375 U.S. 85, 86-87, 84 S.Ct. 229, 11 L.Ed.2d 171 (1963) (search and seizure; “whether there is a reasonable possibility that the evidence complained of might have contributed to the conviction”). Since even the slightest possibility that illegally obtained evidence was used vitiates a conviction, a very effective procedural screen must be interposed.

Most of the cases dealing with burdens of proof on admissibility have been confession cases and the issue has been unconstitutional “coercion.” But it is hard to see why the issue of whether the police “coerced” a confession in violation of the constitution should be treated, for procedural purposes, any differently from that of whether they obtained evidence “as a result of” illegal electronic surveillance. Both words — “coercion” and “result” — have factual and legal content and a finding with respect to the fact component in each presents similar problems to the judge.

Any evidence which is illegally obtained may be damning, so that the decision cannot turn on the importance of a confession in a case. Frequently, having lost on a motion to suppress, the defendant’s chance of avoiding conviction is hopeless. A “confession” may contain exculpatory material, be incomplete, or be unreliable so that it is no more the equivalent of a plea of guilty than is key evidence illegally obtained. Certainly the burden of proof at a preliminary hearing should not depend upon how critical the evidence sought to be suppressed may be at the trial. The judge is often unable to accurately predict at that stage whether a particular item of evidence will clinch the prosecution’s case.

The fact that exclusion in “coercion” cases was historically based upon lack of probative force (Maguire, Evidence of Guilt, 108-109 (1959)) while exclusion of other kinds of illegal evidence has been based upon purely extrinsic policies (People v. Cahan, 44 Cal.2d 434, 439, 442, 445-449, 282 P.2d 905, 907, 909-914 (1955); Maguire, Evidence of Guilt, 167-169 (1959)), does not justify a higher burden in a confession case. In fact, as we have already seen, the probative force argument may be available to the jury where a confession is introduced, while it may not be if evidence obtained by illegal searches or surveillance is used. Cf., as analogous to confession cases, United States v. Davis, 399 F.2d 948, 950, n. 1 (2d Cir. July 17, 1968), (voir dire on identification procedures followed by evidence before jury on reliability of identification in court). There is, therefore, some possibility that the jury will discount a confession which is in fact not voluntary but almost none that it will ignore evidence obtained by illegal electronic surveillance, wiretaps or searches. More, rather than less, protection therefore seems called for in cases such as the one before us.

It is extrinsic policy considerations rather than lack of probative force that now explain many confession exclusion rulings. See, e. g., Spano v. People of the State of New York, 360 U.S. 315, 320-321, 79 S.Ct. 1202, 3 L.Ed.2d 1265 (1959); Jackson v. Denno, 378 U.S. 368, 386, n. 13, 84 S.Ct. 1774, 12 L.Ed.2d 908 (1964). The policies are essentially the same in search and seizure, wiretapping and electronics eavesdropping cases. See, e. g., American Bár Association Project on Minimum Standards for Criminal Justice, Standards Relating to Electronic Surveillance, 114-115 (1968). In each instance an important constitutional right of the defendant is being protected and the policy of deterring future lawlessness by officials is being applied. “The dignity of the United States Government will not permit the conviction of any person on tainted testimony.” Mesarosh v. United States, 352 U.S. 1, 9, 77 S.Ct. 1, 5, 1 L.Ed.2d 1 (1956)

Confessions obtained as a result of violations of constitutional search and seizure requirements are, for that reason alone, inadmissible; “it matters not,” Judge Fuld has properly instructed us, “that these ‘fruits’ happen to be confessions rather than some other type of evidence.” People v. Rodriguez, 11 N.Y.2d 279, 286, 229 N.Y.S.2d 353, 357, 183 N.E.2d 651, 654 (1962). It would make little sense if, as a result of a doubt about a single illegal search, the confession alleged to have been obtained were excluded and the seized real proof were admitted because the burdens were different. Simplicity and ease in administration suggests the desirability of a single high standard in all suppression hearings. There is “no reasonable or logical basis for any distinction between inanimate (tangible) and animate (testimonial) evidence.” People v. Dentine, 21 N.Y.2d 700, 703, 287 N.Y.S.2d 427, 429, 234 N.E.2d 462, 463 (1967) (Fuld, C. J. dissenting).

A workable rule giving adequate consideration to the importance of the exclusionary policy is to require the government to prove beyond a reasonable doubt that the evidence it introduces was legally obtained. This is a familiar standard in criminal cases. It is the one already utilized in some jurisdictions and adopted by the most recent appellate decision on the subject. See Pea v. United States, 397 F.2d 627 (D.C.Cir.1968) and see cases cited at pp. 54, 55, supra.

Application of this standard, based on observations in this Court, would not affect the government adversely in any significant number of cases. Most of the problems with respect to illegally obtained evidence in the federal courts are of an interim nature arising in large part from a lack of clear guidelines available to federal law enforcement personnel rather than to undisciplined lawlessness. As the legal standards become clearer and the procedures for insuring compliance more routinized, motions to suppress because of illegally obtained evidence will become less important than they are today. As a practical matter, even today at hearings on motions to suppress, there is often no serious dispute about the facts (see, e. g., United States v. Feinberg, 383 F.2d 60, 70 (2d Cir. 1967), cert. denied, 389 U.S. 1044, 88 S.Ct. 788, 19 L.Ed.2d 836 (1968))— what Justice Frankfurter has referred to as “the crude historical facts, the external ‘phenonumenological’ occurrences and events.” Culombe v. Connecticut, 367 U.S. 568, 603, 81 S.Ct. 1860, 1879, 6 L.Ed.2d 1037 (1961). Where there remains a reasonable doubt about what happened and whether the government acted lawfully, the evidence should be excluded.

We need not now decide whether this rule ought to apply to factual issues such as consent in search and seizure cases. Cf. United States v. Dawson, 400 F.2d 194 (2d Cir. August 12, 1968). The government’s conduct has concededly been illegal, the information and evidence bearing on the factual questions now before us are peculiarly within its ken, and an important federal policy of protecting citizens against this sort of intrusion on their privacy is at stake. It is fitting that the government bear the burden of proof and that it be a high one. See Pitler, “The Fruit of the Poisonous Tree” Revisited and Shepardized, 56 Calif.L.Rev. 579, 647-48 (1968).

This decision with respect to burdens of proof in suppression cases is consistent with “the high standards which past decisions have insisted be maintained in the conduct of federal criminal trials.” Lawn v. United States, 355 U.S. 339, 365, 78 S.Ct. 311, 326, 2 L.Ed.2d 321 (1958) (concurring opinion) (emphasis supplied). Mesarosh v. United States, 352 U.S. 1, 3, 77 S.Ct. 1, 2, 1 L.Ed.2d 1 (1956) (“the decision herein passes only on the integrity of the criminal trial in the federal courts”). Whether the Constitution requires that the same rule be applied by the states is not a matter we need now consider. Cf. Chapman v. State of California, 386 U.S. 18, 43, n. 2, 87 S.Ct. 824, 17 L.Ed.2d 705 (1967).

(b) Statutory Rule

This analysis of the common law is consistent with the Omnibus Crime Control and Safe Streets Act of 1968. Except for the specific overruling of the McNabb-Mallory rule by subdivision (c) of section 3501 of title 18 of the United States Code (P.L. 90-351, § 701(c)), the language of the Act is directed to the reliability components of the confession-exclusion rules, not to the extrinsic policy components. Senate Re port No. 1097, 1968 U.S.Code Cong. & Admin.News, 1634, 1646-1649.

While Congress was also concerned with the Escobedo-Miranda rules (id. at 1650-1660), the Act itself contains no detailed provisions such as subdivision (c) dealing with these and other constitutional exclusionary rules. Cf. the detailed proposals for dealing with Escobedo-Miranda in American Law Institute, A Model Code of Pre-Arraignment Procedure, Tentative Draft No. 1 (1966); ibid, Study Draft No. 1 (1968); Schaefer, The Suspect and Society, 39-56 (1966).

Such circumstances referred to in subdivision (b) of section 3501 of title 18 as the time between arrest and arraignment, whether the declarant knew the nature of the charge against him, whether he knew of his rights and whether he had the assistance of counsel, may affect the “voluntary” nature of a confession and thus its reliability. Inclusion of these factors in the statute does not foreclose their being considered by the court on separable constitutional issues.

A charge requiring that the jury be satisfied by proof beyond a reasonable doubt of its voluntary nature is apparently not required under the Act. Once the court has passed upon admissibility, if the confession is admitted, it may be given its natural probative force by the jury.

The Act does not address itself to the burden required to be satisfied at the hearing before the judge on the issue of admissibility. Accordingly, that burden remains what it is now, proof beyond a reasonable doubt if exclusion is required because of a violation of defendant’s vital constitutional or statutory rights. (We need not now decide what the burden of proof is on the voir dire where “involuntariness” was not due to official misconduct and only the question of reliability is raised.)

The electronic surveillance by the F.B.I. did not comply with the provisions of Title III of the Omnibus Crime Control and Safe Streets Act of 1968. P.L. 90-351, § 801 et seq., 18 U.S.C. § 2510 et seq. Divulgence at a trial or to other government agents would, accordingly, be prohibited by the Act even if it has retroactive effect. P.L. 90-351, § 802, 18 U.S.C. § 2517.

In short, the Omnibus Crime Control and Safe Streets Act of 1968 has not affected the common law rules applicable to this case.

3. Has the government met its burden?

The government has established by a preponderance of evidence and possibly by clear and convincing proof — though not beyond a reasonable doubt — that the information obtained by the F.B.I. from the electronic surveillance was not furnished to Intelligence. While the New York office of the F.B.I. did ignore, in part, Attorney General Robert F. Kennedy’s insistence on cooperation between all federal law enforcement agencies in the drive against organized crime, there may well have been compliance with the Attorney General’s directive at the highest levels of the F.B.I. in Washington. There is at least a reasonable doubt about whether the F.B.I. reports, which included data obtained through electronic means, were regularly provided to the Organized Crime Division. The Organized Crime Division, in turn, then may have passed this information on to Intelligence through top-level consultations. In this manner, the tainted information might have filtered up from the New York office of the F.B.I. to the Justice Department and then back down to the Intelligence agents in New York.

The only information picked up by the electronic microphones concealed in the Prisco Travel Bureau relevant to defendant’s income tax prosecution concerned possible sources of defendant’s income —his involvement in labor union activities and the carting business and his partial ownership of the Casa Bianca Restaurant. The court, at defendant’s first trial found that “Likely sources for defendant’s income during the indictment years are an interest in a restaurant business [Casa Bianca] and services as a go-between in labor relations of employers with unions.” United States v. Schipani, 63-CR-237, E.D.N. Y., October 15,1965, p. 7.

Two conversations in which the Casa Bianca Restaurant was mentioned were monitored. One was between Clemente and an underworld associate on January 19, 1961 in which it was indicated that Schipani had partial control of the restaurant. The other took place on September 12, 1961 between Clemente and the defendant in which Clemente stated that they should have bought out the Casa Bianca and run it themselves.

This information confirmed what the F.B.I. had strongly suspected back in 1958, namely, that the defendant had an undisclosed interest in this restaurant. In addition, shortly after the Schipani investigation began, the Alcohol Tax Division surveillance indicated that this would be a fruitful line of inquiry. The monitored conversations probably did not furnish the government with any leads with which to pursue this line of proof. All the numerous bits of testimony and evidence relating to the Casa Bianca introduced at the defendant’s trial were obtained through legitimate investigatory techniques.

Nevertheless, the Court finds that the government has not established beyond a reasonable doubt that the evidence of the defendant’s involvement with the Casa Bianca had an independent origin. The information concerning the Casa Bianca obtained through the electronic surveillance would have presumably been furnished to Intelligence shortly after its investigation began at a time when it had to decide which avenues of inquiry were worthy of the limited investigative resources available. This illegally obtained data may have freshened up the government’s legitimate leads and caused it to concentrate more intensively on them.

Nor has the government established beyond a reasonable doubt that the electronic surveillance did not supply leads for its investigation into the defendant’s labor and carting activities in addition to the leads it had lawfully acquired. There were repeated and clear references to these activities in the surveillance logs. Much of the information in the logs could not have been obtained through normal investigative techniques.

The tainted evidence related to specific sources of income — a restaurant, the cartage industry, and labor union activities. Evidence of such sources of income will be suppressed.

Other evidence of defendant’s net worth was proved beyond a reasonable doubt to be untainted. It may be used by the government provided the entire prosecution has not been fatally compromised, an issue to which we now turn.

B. ENTIRE INVESTIGATION

Defendant seeks to suppress all the government’s evidence on the ground that the entire case has been fatally blemished by virtue of the fact that the July 17 order to conduct a saturation investigation of the defendant was a result of the electronic eavesdropping. This contention with respect to the government’s decision to focus on defendant’s activities raises difficult legal and factual issues concerning the scope of the so-called “fruit of the poisonous tree” doctrine and its application to this case.

1. Can a tainted decision to focus abort an entire prosecution?

The legal question presented is whether an entire prosecution can be vitiated if the government obtains information illegally and on the basis of that information decides to throw its weight against an individual, even though the officials in charge of the investigation are unaware of the illicit source of the data and proceed in a lawful manner. Not all evidence, the Supreme Court has recently told us in Wong Sun, “is ‘fruit of the poisonous tree’ simply because it would not have come to light but for the illegal actions of the police.” Wong Sun v. United States, 871 U.S. 471, 488, 83 S.Ct. 407, 417, 9 L.Ed.2d 441 (1963). Rather, the test is “ ‘whether, granting establishment of the primary illegality, the evidence to which the instant objection is made has been come at by exploitation of that illegality or instead by means sufficiently distinguishable to be purged of the primary taint.’ ” Ibid.

If illegally secured information leads the government to substantially intensify an investigation, all evidence subsequently uncovered has automatically “been come at by exploitation of that illegality.” The unlawful search has set in motion the chain of events leading to the government’s evidence. And there has then been “a continuing * * * [and] significant participation by prosecution officials” from the time of the police misconduct to the time of the discovery of the evidence. Ruffin, Out On A Limb Of The Poisonous Tree: The Tainted Witness, 15 U.S.C.A. L.Rev. 32, 38 (1967). “The road from the tap to the testimony may be long, but it is straight.” United States v. Tane, 329 F.2d 848, 853 (2d Cir. 1964). See also Maguire, Evidence of Guilt 246 (1959) (“exclusion of evidence about secondary discoveries made possible solely or facilitated by the original violation”); PitIer, “The Fruit of the Poisonous Tree” Revisited and Shepardized, 56 Calif.L.Rev. 579, 642-46 (1968).

In the present case, if, as a result of the electronic eavesdropping, the government overheard Schipani say “I robbed the ‘X’ bank” and then sought to prosecute him for that bank robbery after an investigation touched off by this remark, it is doubtful if the government could successfully withstand a motion to dismiss the indictment. See Smith v. United States, 120 U.S.App.D.C. 31, 344 F.2d 545 (1965) (discovery that crime had been committed as a result of illegal search); United States v. Tane, 329 F.2d 848 (2d Cir. 1964) (government first learns from wiretapped conversation that defendant was receiving unlawful payments); United States v. Coplon, 185 F.2d 629 (2d Cir. 1950) (if police “had been set on the [defendant’s] trail” by wiretapping, entire prosecution vitiated); People v. Dentine, 21 N.Y.2d 700, 287 N.Y.S.2d 427, 428-430, 234 N.E.2d 462 (1967) (Fuld, J., dissenting); cf. United States v. Hagarty, 388 F.2d 713 (7th Cir. 1968) (electronic surveillance). See also Broeder, Wong Sun — A Study in Faith and Hope, 42 Neb.L.Rev. 483, 523 (1962). Similarly, if the government sought to rely upon the testimony of a witness whose identity was revealed by an unlawful source, it would not be permitted to do so. See, e. g., People v. Martin, 382 Ill. 192, 46 N.E.2d 997 (1942).

The result should not be different if the government learns through illegal means that a person is closely connected with organized crime and then proceeds to launch an income tax investigation. Discovery of such information has much the same effect as overhearing the defendant say “I haven’t paid taxes for the last five years.” The fact that a general exploratory investigation of the defendant is already in progress at the time does not license investigators to use illegal leads. Cf. Williams v. United States, 382 F.2d 48 (5th Cir. 1967); People v. Dentine, 21 N.Y.2d 700, 287 N.Y.S.2d 427, 430, 234 N.E.2d 462 (1967) (Fuld, J., dissenting); People v. Mills, 148 Cal.2d 392, 306 P.2d 1005, cert. denied, 355 U.S. 841, 78 S.Ct. 55, 2 L.Ed.2d 46 (1957).

The unique circumstances of an income tax investigation make a decision to focus intensively of critical importance. As opposed to crimes like assault or robbery, tax evasion is hidden. There are at least hundreds of thousands of tax violators whose criminality has not been revealed. One of the chief problems for the government is to decide how it is going to utilize its limited tax investigation forces. The main hope of a tax violator is that the Internal Revenue Service will remain unaware of his existence. Once the government begins to concentrate all its enormous resources on a citizen, the chance of its discovering that he has violated the tax laws is greatly multiplied. It is difficult to perceive how the government could receive any more valuable information than the name of a probable tax violator.

We recognize that this approach to the focusing decision may be viewed as contrary to that reached by Judge Learned Hand in the 1942 case of United States v. Nardone, 127 F.2d 521 (2d Cir.), cert. denied, 316 U.S. 698, 62 S.Ct. 1296, 86 L.Ed. 1767 (1942). In that case, Judge Hand declined to apply the “fruit of the poisonous tree” doctrine where the information obtained by wiretapping admittedly “spurred the authorities to press an investigation which they might otherwise have dropped.” 127 F.2d at 523. He declared that “the Supreme Court [had not] meant to involve the prosecution of crime in such a tenebrous and uncertain inquiry, or make such a fetich of the statute as so extreme an application of it would demand.” Ibid.

The Nardone case is distinguishable for two reasons. First, the unique circumstances surrounding an income tax prosecution, noted above, serve to distinguish Nardone — an alcohol smuggling prosecution — from the instant case.

Second, it has been undermined by recent decisions of the Supreme Court and lower federal courts. The test announced in Wong Sun, which focuses on police exploitation of the illegality, significantly expands the area which can be tainted by an unlawful search and seizure. See, e. g., Smith v. United States, 120 U.S.App.D.C. 160, 344 F.2d 545 (1965); United States v. Tane, 329 F.2d 848 (2d Cir. 1964); cf. Harrison v. United States, 392 U.S. 219, 224, 88 S.Ct. 2008, 2011, 20 L.Ed.2d 1047 (1968) (question whether “petitioner’s trial testimony was * * * impelled by * * * wrongful use of his illegally obtained confessions”). The courts have within recent years felt it their duty to engage in what Judge Hand would have described as “tenebrous and uncertain inquiry” where constitutional rights are involved. See Harrison v. United States, 392 U.S. 219, 88 S.Ct. 2008, 20 L.Ed.2d 1047 (1968); Smith v. United States, 120 U.S.App.D.C. 160, 344 F.2d 545 (1965) (“by contacting Donaldson, the police created the pressure which impelled Hardy to turn the transmission over to the police”).

In Harrison v. United States, for example, the prosecution introduced petitioner’s testimony at a prior trial in which he had denied committing the offense in question, but admitted being at the scene; this testimony had been given after petitioner’s confession — later ruled to be inadmissible — had been introduced. The Supreme Court reversed on the ground that the government had not demonstrated “that the petitioner’s testimony was obtained ‘by means sufficiently distinguishable’ from the underlying legality ‘to be purged of the primary taint’ ”. 392 U.S. at 226, 88 S.Ct. at 2012. The question, the court declared,

“is not whether the petitioner made a knowing decision to testify, but why. If he did so in order to overcome the impact of confessions illegally obtained and hence improperly introduced, then his testimony was tainted.” 88 S.Ct. at 2010 (emphasis in original).

The Supreme Court felt it necessary to inquire into the motives of a defendant to testify at his trial. Federal courts cannot avoid an inquiry to determine whether the intensity of the government’s investigation of a possible tax violator was substantially affected by illegally elicited information.

2. What is the government’s burden of proof on the focusing issue?

The exacting standard of beyond a reasonable doubt is not appropri ate to the issue of whether the intensity of the government’s total investigation was substantially affected by illegal electronic surveillance. Rather, we think, a preponderance of the evidence test is called for.

First, we are not dealing with an item of evidence that will ultimately be introduced at a trial and relied upon by the jury. Thus, the consideration that the defendant be convicted beyond a reasonable doubt on legally obtained evidence is not as important as it is when individual items of evidence are considered.

Second, the courts should avoid wherever possible making detailed inquiries into the reasons for executive decisions and examining communications among high officials which are the basis of their exercise of judgment. Cf. 8 Wigmore, § 2378(3) (McNaughton rev. 1961); Gellhorn and Byse, Administrative Law Cases and Comments, 617-618 (4th ed. 1960). Not only is this reluctance supported by proper respect for the separation of powers, but by the practical consideration that it is often impossible for a highly placed official to analyze or remember the varied emotional and intellectual reasons for a decision. A decision to intensify an investigation is one of executive judgment and it would be unrealistic to require the government to meet a higher burden of proof than a preponderance.

3.' Has the ■ government met its burden?

The government has established by a preponderance of evidence and possibly by clear and convincing proof— though not beyond a reasonable doubt- — ■ that, the intensity of its investigation by Intelligence and the Alcohol Tax Division was not affected by tainted information.

It is highly likely that the government would have launched a saturation investigation of the defendant even if no information had been received from electronic surveillance of the Prisco Travel Bureau. The untainted F.B.I. report of October 10, 1960 identified the defendant as a powerful underworld figure, associated with many of New York’s top mobsters. He was reputed to have “inherited” the gambling interests of Joe Adonis and to have undisclosed interests in three restaurants.

The investigation touched off by Attorney General Robert F. Kennedy’s organized crime drive soon brought additional leads to light through acceptable investigative techniques. Defendant had never filed an income tax return although he obviously enjoyed a high standard of living. Immediately after the investigation began, the defendant appeared so vulnerable and the likelihood that an income tax investigation would prove fruitful was so great, that government agents would naturally focus on him.

This strong scent of big game probably explains why Intelligence Special Agent Masetti’s unilateral decision to switch to an income tax investigation was acquiesced in by his superiors in Washington. Masetti’s expertise in income tax fraud is a natural and reasonable explanation of his own decision. While there does remain a reasonable doubt about the motives for the switch from a limited series of investigations on narrow grounds — wagering tax and alcohol tax forms — to a broad scale income tax investigation, the government has proved by a preponderance that this decision was not based on illegally obtained evidence. The government’s burden on this issue has been satisfied.

The Alcohol Tax Division’s- surveillance, which also resulted from the January, 1961 decision to begin a preliminary investigation, revealed that the defendant was driving a car registered in another person’s name, paying bills in cash, and, of crucial importance, purchasing large quantities of money orders. This latter lead was first discovered on July 17, 1961 — the same date on which Intelligence was requested to begin a saturation investigation. These initial lawfully made discoveries fed on each other. As more leads were uncovered, the intensity of the investigation steadily escalated, leading to the uncovering of still more evidence.

The information elicited through electronic surveillance did not significantly contribute to the progress of the investigation or encourage the investigators. Most of the information obtained from this source was already known. Only one F.B.I. report on Schipani was filed between the time the electronic surveillance began and the time the decision was made to launch a saturation investigation. This report, dated June 15, 1961, revealed the defendant’s close association with Clemente, his plans to open a carting business, and his involvement in the sale of Puerto Rican sweepstake tickets on the New York docks. In addition, the suspicion that Schipani had a partial interest in the Casa Bianca was confirmed. Although the specific information concerning Schipani’s plan to form a carting company was new, his position as a major figure in the Brooklyn carting industry had been known for some time. Possibly, it is this information, plus that furnished by the Alcohol Tax Division, to which the July 17 saturation order refers when mention is made of “information supplied * * * by other investigative agencies.” It cannot be said that these comparatively minor leads — as opposed to those uncovered through legitimate investigatory techniques by Intelligence and the Alcohol Tax Division — were responsible for the increase in the intensity of the Schipani investigation.

CONCLUSION

All evidence relating to the defendant’s likely sources of income which was introduced at the first trial is suppressed. The motion is denied in all other respects.

So ordered.