United States v. Lopez

MEMORANDUM

Defendant is charged with concealing and facilitating the transportation of heroin and of conspiring to commit that crime. 21 U.S.C. §§ 173, 174. He has moved to suppress heroin taken from his person.

Presented is the question of whether the anti-hijacking system used at our airports is constitutional. For the reasons indicated below, we hold the system valid. Nevertheless, because in this particular case the elegant and objective method devised by the government to deter and apprehend hijackers was distorted in an irrational and prejudicial way by airlines personnel, defendant’s motion must be granted.

I. FACTS

With his companion, Ernesto Perez Gonzalez, defendant, Frank Lorenzi Lopez, was apprehended on November 14, 1970 at John F. Kennedy International Airport as he was about to board a Pan American flight bound for Puerto Rico. Two Deputy United States Marshals had been called from a trailer adjacent to the Pan American premises where they were stationed in connection with the government’s anti-hijacking program. An employee of Pan American had pointed out the two passengers as “selectees” *1082 —persons whose “profile” suggested a substantial likelihood that they were potential hijackers; who had activated a magnetometer, a metal detection device; and who, upon request, had failed to produce identification.

The Marshals approached the two and asked if they would walk through the magnetometer installation again, first with, and then without, a small blue bag that Gonzalez was carrying. Each of them did so, activating the device on both trips. The Marshals again requested identification but none was produced. At this time Gonzalez identified himself by his proper name, indicating that the name Julio Lopez, which appeared on his ticket, was erroneous.

The two travelers were then asked to accompany the Marshals to a private area adjacent to the passenger boarding ramp where their outer clothing was patted down (“frisked”) for weapons. A Marshal felt a hard object about 4 inches wide, 6 inches long, and three-quarters of an inch deep under Lopez’s clothing. A tinfoil covered plastic envelope tightly packed with white powder had been discovered. Field tested, the powder proved positive for heroin. Gonzalez and Lopez were then arrested.

Both were charged with concealment and facilitating the transportation of the packet of narcotics found on Lopez and with conspiring to commit this crime. The government’s evidence at the suppression hearing made it apparent that there was no case against Gonzalez, even if the heroin was admitted. Accordingly, a judgment of acquittal was granted after the government and Gonzalez waived a jury and agreed that the hearing testimony would be deemed received at the trial.

II

FEDERAL AERONAUTICS ADMINISTRATION SYSTEM FOR DISCOURAGING AND APPREHENDING POTENTIAL HIJACKERS

In October 1968 a Task Force was appointed to consider methods of combating the increasing number of airline hijackings. Cf. Bibliography on Airplane Hijacking, 26 The Record of the Assoc. of the Bar of the City of New York, 325-332 (1971). A number of interested agencies including the Federal Aeronautics Administration, the Department of Justice and the Department of Commerce were represented. This Task Force included individuals trained in several disciplines including psychology, law, engineering and administration. Dr. John T. Dailey, a well-trained psychologist with a broad practical background in education and government personnel, took a leading role in developing and testing the Task Force ideas. At the hearing he testified at length and impressed the Court with his skill and honesty.

One of the serious problems faced by the Task Force was that many millions of passengers use air transportation. Any practical procedure would have to permit maximum access to aircraft with minimal inconvenience and embarrassment to passengers and almost no delay in the operations of the airlines.

Among the investigations undertaken by this group were a detailed study of the characteristics of all the then known hijackers and of the air traveling public. Background investigations of hijackers as well as visual and photographic studies of boarding air passengers were relied upon. Among the findings were (1) hijackers were generally not highly motivated and resourceful and (2) they shared certain characteristics markedly distinguishing them from the general traveling public. In addition, engineering studies were undertaken of available weapon detection devices.

After thorough field testing of equipment and various screening techniques, the present anti-hijacking system was instituted. Although we are concerned primarily with the preflight apprehension aspect of the system its principal *1083 focus is on deterrence. Involved are the following elements:

1. Heavy penalties.

A severe statute with possible death penalties for hijacking was adopted. There already were in force various regulations and statutes prohibiting the carrying on board of firearms and other weapons.

2. Notice to the Public.

Signs in English and Spanish are posted at the boarding gates where passengers’ tickets are checked reading as follows:

AIRCRAFT HIJACKING IS A FEDERAL CRIME PUNISHABLE BY DEATH

CARRYING CONCEALED WEAPONS ABOARD AIRCRAFT IS PUNISHABLE BY PRISON SENTENCES & FINES

PASSENGERS AND BAGGAGE SUBJECT TO SEARCH

These signs are eleven by fourteen inches with half inch high letters. Warnings of the same or larger dimensions are conspicuously posted at other parts of the air terminal.

While probably not required to give notice of the applicable law and penalties, these signs fill that function. Cf. Lambert v. California, 355 U.S. 225, 228-230, 78 S.Ct. 240, 242-243, 2 L.Ed.2d 228 (1957) (felon registration ordinance); United States v. Mancuso, 420 F.2d 556 (2d Cir. 1970) (failure to register as a narcotics law violator before leaving the country). They serve to deter and to reduce the possibility of embarrassment should a passenger’s boarding progress be interrupted.

3. Profile.

If a passenger meets a prescribed “profile” he is focused on by airline employees. The details of the profile and its use are set out below.

4. Magnetometer.

A magnetometer is installed in the passageway leading to the plane so that all passengers must pass through it. It is set to flash a warning light when metal equal to or greater than an average 25 caliber gun in magnetic force deflecting power is carried by. This device is described in more detail below.

5. Interview by Airlines Personnel.

A person who triggers the magnetometer and meets the profile requirements is “interviewed” by airlines personnel. If he provides satisfactory identification, he is permitted to proceed unimpeded. Otherwise he is designated a “selectee” and is denied boarding until a Deputy United States Marshal is summoned.

6. Interview by Marshal.

The Marshal again requests identification of those designated as “selectees.” If satisfactory identification is not furnished it is suggested that the person go through the magnetometer once more. Before walking through, he is asked if he has any metal on his person or in any baggage he is carrying. If he replies in the negative and still sets off the magnetometer, a request is made that he submit to a “voluntary” search. It is explained that this search is part of an attempt by the government to prevent hijacking.

7. Frisk.

The Marshal pats-down the external clothing of the subject in order to discover if he is carrying any weapons. Depending upon what is found as a result of the frisk, boarding is permitted or the person,is detained.

The program is designed to speed passengers who are unlikely to present danger and to isolate, with the least possible discomforture or delay, those presenting a substantial probability of danger. At each successive screening stage an attempt is made to permit as many as possible to complete boarding.

While no single screening technique can by itself completely protect the flying public — without creating an objectionable level of disturbance and inconvenience — probabilities are increased by combining several approaches, thus sufficiently reducing the size of the population which must ultimately be physically *1084 interfered with to a practicable and socially acceptable level. Cf. Rosado v. Wyman, 322 F.Supp. 1173, 1180-1181 (E.D.N.Y.), aff’d, 437 F.2d 619 (1970) (collecting authorities on use of mathematics in the law); Finkelstein and Fairley, A Bayesian Approach to Identification Evidence, 83 Harv.L.Rev. 489 (1970).

The system seems to serve this purpose well. Whether because of it or for other reasons hijacking decreased in 1970 to approximately 50% of what it was in 1969. No flight fully protected by the program has been hijacked.

One sample consisting of 500,000 screened passengers showed^ that only 1,-406 satisfied the profile' — '.28%. Approximately one-half of those were nevertheless permitted to board immediately after failing to activate the magnetometer, leaving 712, or .14% to be interviewed. Of those interviewed, 283, approximately one-third, were actually searched. Therefore, only .05% of the sample were ultimately subjected to a preventive weapons frisk. Twenty persons were denied boarding — approximately Vis of those searched and of these, 16 were arrested. In sum, almost everyone (99.86%) of the one-half million persons passed swiftly through the boarding process without even being asked a question and 99.95% boarded without being searched.

In another sample of 226,000 screened passengers .57% were selected as meeting the profile; .28% were interviewed; and .13% were searched. It was reported that none were searched “involuntarily” and only 24 were denied boarding.

Statistics for a number of covered airlines at the John F. Kennedy International Airport in December 1970 were produced. While the number of passengers using these airlines during this month is not shown, they carried a total of 2,645,-000 passengers during the last six months of 1970 — or on the average 441,000 each month. Three hundred and three reportedly met the profile and nine arrests were made. There were confiscations of 10 illegal knives, one tear gas pen, four hand guns, one gas gun and one pack of marijuana. For the reasons described in part VIII, infra, of this opinion, the Kennedy statistics, insofar as they are affected by Pan American Airways figures, cannot now be relied upon.

That the risk of hijacking is greatly increased when a passenger possesses weapons can hardly be doubted. In the 80 hijacking incidents involving planes of United States registry up to June of 1970, there were 55 firearms, 20 knives, 14 alleged bombs, 3 razors or razor blades, 1 BB gun, 1 tear gas pen, and 1 broken bottle.

In camera testimony was persuasive that the characteristics of the potential hijackers chosen for the profile were well calculated to eliminate safe persons while isolating those likely to be dangerous. No one can be certain, of course, that anyone failing all the tests of the system will be a hijacker. In fact, approximately 14 out of every 15 people who were searched proved to have no weapons and were then permitted to proceed. Moreover, as the facts detailed below indicate, there is always the risk that a soundly designed process will be abused by ignorant, careless or malevolent personnel. As one commentator pointed out

“permitting any use of certain mathematical methods entails a sufficiently high risk of misuse, or a risk of misuse sufficiently costly to avoid, that it would be irrational not to take such misuse into account when deciding whether to permit the methods to be employed at all.” Tribe, Trial by Mathematics: Precision and Ritual in the Legal Process, 84 Harv.L.Rev. 1329, 1331 (1971) (emphasis in original).

Measured against the air traveling population as a whole, the method is highly effective in narrowing the group which needs particular attention. Where the risks of hijacking to passengers and crew and to the viability of the entire industry are so great we cannot say on balance that use of the system is imprudent. Whether it meets the reasonableness test of the Constitution is an issue dealt with below.

*1085 Before turning to that problem there are two procedural difficulties that need to be resolved. One of them, the issue of in camera hearing, raises very serious questions. The other, acceptance of proof that the magnetometer did what it was purported to have done, is relatively simple. We turn to the simple issue first.

Ill

MAGNETOMETER .

The electronic weapons detector — appropriately named “Friskem” — utilized in this case depends upon magnetic field detectors called “flux-gate magnetometers.” The unit was developed after conferences with representatives of the Task Force.

Its operation is based upon the physical fact that the earth is surrounded by a relatively constant magnetic field composed of lines of flux. Steel and other ferromagnetic metals are much better conductors than the air. As a result, when any such metal moves through an area, nearby magnetic lines of flux are distorted to some degree as they tend to converge and pass through the metal while seeking the path of least resistance. Such distortions occurring near a “flux-gate magnetometer” create a signal which can be amplified and calibrated to detect magnetic disturbances. See, e. g., Chapman, The Earth’s Magnetism, 10-12, 17-19, 27, 28 (2d ed. 1951); J. Jaquet, No-Touch Frisk Electronic Weapons Detection paper presented at Conference on Electronic Crime Countermeasures, U. of Ky., April 22, 1971; Marshall, An Analytic Model for the Fluxgate Magnetometer, IEEE Transactions on Magnetics, Vol. MAG-3, No. 3 (Sept. 1967); Geyger, Flux-Gate Magnetometer Uses Toroidal Core, Electronics (June 1, 1962); Geyger, The Ring-Core Magnetometer— A New Type of Second-Harmonic Flux-Gate Magnetometer, Communication and Electronics (Mar. 1962).

Though these scientific principals are not matters of common knowledge they may be readily and accurately determined, are verifiable to almost a certainty and are not disputed. The literature was placed in the Court file, notice was given to the parties that the Court intended to rely upon it, and there was no objection by either party.

Under these circumstances the Court takes judicial notice of the scientific principles utilized in the design of the Friskem unit. It finds that such a machine, if properly constructed and operated, can perform in the manner described to the Court by testimony and manuals. See, e. g., Revised Proposed Rules of Evidence for the United States District Courts and Magistrates, Rule 201(b) (2) (March 1971); Korn, Law, Fact, and Science in the Courts, 66 Colum.L.Rev. 1080, 1089, 1107-1115 (1966); Davis, Judicial Notice, 55 Colum.L.Rev. 945, 948-952 (1955). Cf. Application of Hartop, 311 F.2d 249, 255-257, 50 CCPA 780 (1962) (safety of drugs); Golaris v. Jewel Tea Co., 22 F.R.D. 16, 20 (N.D.Ill.1958) (trichinosis dangers); United States v. Dreos, 156 F.Supp. 200, 208 (D.Md.1957) (radar); The S.C.L. No. 9, 37 F.Supp. 386, 391 (E.D.Pa.1939), aff’d, 114 F.2d 964 (3d Cir. 1940) (bouyancy); State v. Tomanelli, 153 Conn. 365, 216 A.2d 625 (1966) (radar); McKay v. State, 155 Tex.Cr.R. 416, 235 S.W.2d 173 (1950) (drunkometer); State v. Damm, 64 S.D. 309, 266 N.W. 667 (1936) (blood type).

Since no opinion was brought to the Court’s attention taking judicial notice of magnetometer capabilities, the Court also relied upon expert testimony adduced at the hearing. Such reliance is often the first step in a process that passes through judicial notice to acceptance on a theory of stare decisis. See, e. g., Maguire, et al., Cases and Materials on Evidence, 22-29, 65 (5th ed. 1965). This has been the experience in connection with such scientific techniques as use of fingerprints, ballistic comparison and radar. Testimony and exhibits adduced at the hearing and trial indicated that the Friskem device was designed, in accordance with scientific principles already described, specifically to meet the *1086 requirements of the Task Force and that it was properly manufactured.

The model used in the instant case consists of a series of magnetometers in two vertical poles located 36 inches apart and amplification equipment and indicators contained in a console unit. As demonstrated in the Court, and as revealed in the various manuals, the operation of the equipment requires no understanding of its theory. Its calibration is easy and its adjustment can be assured by simple visual observations.

While the person who actually adjusted the machine at the airport was not available, it was proper to accept testimony with respect to the custom and practice of the enterprise. See Revised Proposed Rules of Evidence for the United States District Courts and Magistrates, Rule 406 (March 1971). That practice demonstrated that the physical equipment was adjusted so that a light would flash when metal having the effect of a 25 caliber pistol passed through it.

The Friskem unit will be activated by a variety of objects including a collection of ordinary belongings carried by passengers. In fact, somewhere in the order of 50% of the persons who pass through trigger it. Since, however, this is only one of a series of separating stages intended as much for deterrent value as for ability to ferret out metal objects, its usefulness is substantial.

IV

IN CAMERA INVESTIGATION OF PROFILE

Were even one characteristic of the “profile” generally revealed, the system could be seriously undermined by hijackers fabricating an acceptable profile. Because of this serious danger, the Court excluded from the courtroom the public and the defendant — but not his attorney —when it took testimony which revealed the specific characteristics included in the profile and the findings which led to its adoption.

Defense counsel were permitted to hear all the testimony about the specific characteristics and were able to cross-examine witnesses. They were informed that should it be necessary to consult with their clients about any of this information they could make this known to the Court and it would then reconsider its decision to deny the information to the defendants. No such request was made.

The Court enjoined counsel not to reveal the profile and it has confidence that its injunction will .be followed. Defendant" árgued that he be permitted to attend and be similarly enjoined but such an injunction could never be enforced. The evidence revealed that he was a narcotic addict or a dealer in narcotics or both who would feel no compunction about telling what he knew to all who would lend an ear in prison or out. Under these circumstances the Court had no choice but to exclude him.

The testimony revealed that studies underlying the profile were thorough. Procedures followed in developing it were adequate. Appropriate statistical, sociological and psychological data and techniques were utilized. The profile is a highly effective procedure for isolating potential hijackers.

After studying known hijackers, the task force compiled twenty-five to thirty characteristics in which hijackers differed significantly from the air-traveling public. By putting only a few of them together they could obtain a reliable combination sharply differentiating potential hijackers from non-hijackers.

During the testing period and shortly thereafter in 1969 the task force studied an additional sample of 30 new hijackers and found that over 90% of that group would have met the profile. There has been a continuous process of reevaluation in light of new hijackings and changes in trends of hijacking. Thus far the original profile has retained validity.

Those characteristics selected can be easily observed without exercising judgment. They do not discriminate against any group on the basis of religion, origin, political views, or race. *1087 They are precisely designed to select only those who present a high probability of being dangerous. Thus, they violate none of the traditional equal protection standards. See Williams v. Rhodes, 393 U.S. 23, 89 S.Ct. 5, 21 L.Ed.2d 24 (1968) (discrimination against new parties); Loving v. Virginia, 388 U.S. 1, 87 S.Ct. 1817, 18 L.Ed.2d 1010 (1967) (discrimination against race in miscegenation); Harper v. Virginia State Board of Elections, 383 U.S. 663, 86 S.Ct. 1079, 16 L.Ed.2d 169 (1966) (discrimination against poor voters); Yick Wo v. Hopkins, 118 U.S. 356, 6 S.Ct. 1064, 30 L.Ed. 220 (1886) (discrimination against race in licensing); cf. Bolling v. Sharpe, 347 U.S. 497, 74 S.Ct. 693, 98 L.Ed. 884 (1954) (violations of equal protection standards are also violations of due process under Fifth Amendment).

The constitutional issue raised by the decision to exclude the defendant from the inquiry on the profile, while permitting his counsel to attend and cross-examine witnesses, is obvious. Defendant has both a right to confront witnesses against him and to a public trial.

(a) Public trial

A public trial is guaranteed by the Fifth and Sixth Amendments. See also Fed.R.Crim.P. 43; Annot., Accused’s Right, Under Federal Constitution, to be Present at This Trial, 25 L.Ed.2d 931 (1970). As the Supreme Court noted:

“We start with the proposition that it is a ‘public trial’ that the Sixth Amendment guarantees to the ‘accused.’ The purpose of the requirement of a public trial was to guarantee that the accused would be fairly dealt with and not unjustly condemned.” Estes v. Texas, 381 U.S. 532, 538-539, 85 S.Ct. 1628, 1631, 14 L.Ed.2d 543 (1965).

American distrust for secret proceedings was roused by persecutions of the Inquisition and the Star Chamber; it has been nurtured by contemporary examples of abuse in a variety of repressive regimes. “The knowledge that every criminal trial is subject to contemporaneous review in the forum of public opinion is an effective restraint on possible abuse of judicial power.” In re Oliver, 333 U.S. 257, 268-272, 68 S.Ct. 499, 505, 507, 92 L.Ed. 682 (1948). “[S]ecret tribunals were effective instruments of oppression.” Estes v. Texas, 381 U.S. 532, 539, 85 S.Ct. 1628, 1631, 14 L.Ed.2d 543 (1965). This right to a public trial attaches at the suppression hearing — often the crucial stage. United States ex rel. Bennett v. Rundle, 419 F.2d 599, 605-608 (3d Cir. 1969).

The public has an independent right to be present to see that justice is fairly done. It is important that our citizens be free to observe court proceedings to insure a sense of confidence in the judicial process. Conducting trials behind closed doors might engender an apprehension and distrust of the legal system which would, in the end, destroy its ability to peacefully settle disputes. In re Oliver, 333 U.S. 257, 270, 68 S.Ct. 499, 506, n. 24, 92 L.Ed. 682 (1948); Lewis v. Peyton, 352 F.2d 791, 792 (4th Cir. 1965); cf. Estes v. Texas, 381 U.S. 532, 541, 85 S.Ct. 1628, 1632, 14 L.Ed.2d 543 (1965); United States ex rel. Bennett v. Rundle, 419 F.2d 599, 606 (3d Cir. 1969).

Despite the important rights conferred on both the defendant and the public by the Constitution, exclusion of the public, or portions of it, for limited purposes and for short periods is sometimes justified in the public interest or in the interest of the defendant. Thus the risk of danger to witnesses or the need to preserve order warrants closed proceedings. See, e. g., Sheppard v. Maxwell, 384 U.S. 333, 86 S.Ct. 1507, 16 L.Ed.2d 600 (1966) (authorizing limitation of the press and media during the trial to protect the defendant); Estes v. Texas, 381 U.S. 532, 85 S.Ct. 1628, 14 L.Ed.2d 543 (1965) (same); United States ex rel. Bruno v. Herold, 408 F.2d 125 (2nd Cir. 1969), cert. denied, 397 U.S. 957, 90 S.Ct. 947, 25 L.Ed.2d 141 (1970) (some spectators removed to avoid harassment of witness); United States ex rel. Orlando v. Fay, 350 F.2d 967 (2d Cir. 1965), cert. denied, Orlando v. Follette, 384 U.S. 1008, *1088 86 S.Ct. 1961, 16 L.Ed.2d 1021 (1966) (all spectators but press removed to preserve order); Geise v. United States, 262 F.2d 151 (9th Cir. 1958), cert. denied, 361 U.S. 842, 80 S.Ct. 94, 4 L.Ed.2d 80 (1959) (most of spectators cleared in rape case where prosecutrix and two other witnesses were of tender years and a large audience would inhibit testimony); Melanson v. O’Brien, 191 F.2d 963 (1st Cir. 1951) (general public excluded by state law in sex crime case where victim was a minor).

The danger in revealing the profile is so great as to warrant the public’s exclusion for a limited period. There was no violation of the right to a public trial.

(b) Right of confrontation

A much more difficult issue is presented by exclusion of the defendant himself. Normally vital to a fair trial are Sixth Amendment rights to confront witnesses. Dutton v. Evans, 400 U.S. 74, 91 S.Ct. 210, 215, 27 L.Ed.2d 213 (1970); California v. Green, 399 U.S. 149, 90 S.Ct. 1930, 26 L.Ed.2d 489 (1970). So, too, are the more general elements of fairness in due process. Illinois v. Allen, 397 U.S. 337, 338, 90 S.Ct. 1057, 1058, 25 L.Ed.2d 353 (1970); Snyder v. Massachusetts, 291 U.S. 97, 106, 54 S.Ct. 330, 332, 78 L.Ed. 674 (1934), overruled on other grounds, Malloy v. Hogan, 378 U.S. 1, 84 S.Ct. 1489, 12 L.Ed.2d 653 (1964); Lewis v. United States, 146 U.S. 370, 372-373, 13 S.Ct. 136, 137, 36 L.Ed. 1011 (1892); Fed.R.Crim.P. 43. As in the case of the right to a public trial, a suppression hearing is considered part of the trial. United States v. Dalli, 424 F.2d 45, 48 (2d Cir.), cert. denied, 400 U.S. 821, 91 S.Ct. 39, 27 L.Ed.2d 49 (1970). See Lewis v. United States, 146 U.S. 370, 13 S.Ct. 136, 36 L.Ed. 1011 (1892) (impaneling of jury); United States v. Crutcher, 405 F.2d 239 (2d Cir. 1968), cert. denied, 394 U.S. 908, 89 S.Ct. 1018, 22 L.Ed.2d 219 (1969) (same).

The right is not absolute. It may be waived by acts or statements of the defendant. Diaz v. United States, 223 U.S. 442, 32 S.Ct. 250, 56 L.Ed. 500 (1912); United States v. Dalli, 424 F.2d 45 (2d Cir.), cert. denied, 400 U.S. 821, 91 S.Ct. 39, 27 L.Ed.2d 49 (1970); United States v. Crutcher, 405 F.2d 239 (2d Cir. 1968), cert. denied, 394 U.S. 908, 89 S.Ct. 1019, 22 L.Ed.2d 219 (1969). The Court may also exclude the defendant when necessary to preserve order and decorum in the courtroom. Illinois v. Allen, 397 U.S. 337, 90 S.Ct. 1057, 25 L.Ed.2d 353 (1970).

It might be argued that exclusion of the defendant constitutes harmless error when his counsel was present and the defendant could not by his presence have contributed to his defense. In this case the only in camera testimony concerned revelation of the actual criteria contained in the profile which the defendant irrefutably satisfied. Defense counsel was present throughout the testimony and cross-examined the witnesses thoroughly. Arguably no prejudice resulted. See Chapman v. California, 386 U.S. 18, 87 S.Ct. 824, 17 L.Ed.2d 705 (1967). The harmless error argument is, however, not sufficiently persuasive to justify the procedure adopted.

It is true that this defendant was in no position to assist his attorney with detailed knowledge of the events essential for a variety of tactical and other decisions in cross-examining, objecting and the like. See Snyder v. Massachusetts, 291 U.S. 97, 54 S.Ct. 330, 78 L.Ed. 674 (1934), overruled on other grounds, Malloy v. Hogan, 378 U.S. 1, 84 S.Ct. 1489, 12 L.Ed.2d 653 (1964); Lewis v. United States, 146 U.S. 370, 13 S.Ct. 136, 36 L.Ed. 1011 (1892). But any defendant, if he is found guilty, should, as a matter of fundamental fairness and as part of the rehabilitative process, have the assurance, by reason of his direct observation, that justice was done. There could hardly be anything more rankling to a defend