I. Introduction

As Justice Brandeis so eloquently argued more than a century ago in his classic privacy article, the progress of science, especially in the area of communication technology, made it imperative that we shift our attention from the letter to the spirit of law to protect the individual against the privacy invasions of modern inventions. [1] The technological landscape continues to change at a much faster pace than constitutional law, which has yet to deal with such new contrivances of science as computer communications across telephone lines and electronic surveillance of the public by sophisticated devices that detect, monitor, and record personal information. The purpose of this Article is to explore and determine the constitutionality of electronic mail searching and scanning by government in light of recent Fourth Amendment rulings and trends.

Toward this end, I will begin by presenting some of the new modes of communication, and proceed to assess the constitutionality of the warrantless search and seizure of electronic mail. Since this issue has not been addressed by the U.S. Supreme Court yet, I will attempt to develop [p136] what is arguably a proper basis for constitutional analysis regarding this emerging area of Fourth Amendment search law by relying on federal and military court decisions as well as the valuable insights of respected commentators. I will also point out the situations in which Fifth Amendment issues could potentially arise. In conclusion, I will answer some hypothetical questions and propose some guidelines for determining when government action is unjustifiably intrusive, because what is constitutionally permissible may not necessarily be a good public policy.

II. Modes of Modern Communication

A. Land Line Telephones

Although land line telephones, which date back to the last quarter of the nineteenth century, can hardly be described as modern anymore, they are probably the best departure point in our inquiry for two reasons. First, many of the subsequent instruments of telecommunication networks, such as teleprinter exchange (telex), facsimile transmissions (faxes), and electronic mail (e-mail), are essentially based on the telephone and make use of the same underground cables, digital lines, radios, and satellite links to make connections between two or more users. Second, telephones have been in use by the police and the public long enough to allow the development of a sizable body of legal code and case law around them. The constitutional analysis of these cases may be extended to the newer and more advanced forms of communication technology.

In the seminal case of Katz v. United States, [2] the Court found that the warrantless wiretapping of standard land line telephones constituted an unreasonable search. This means that land line telephone calls are constitutionally protected against warrantless seizure and subsequent use in a criminal trial by government, because a person placing a call “may rely upon the protection of the Fourth Amendment,” whether he happens to be in a business office, a friend’s apartment, a taxicab, or a telephone booth. [3] Before Katz, postal mail was the only commonly used communication medium protected from unauthorized interception. [4] Most professionals were then forced to meet with their clients in person or exchange information with them via mail in order to protect their confidences and, [p137] in the case of legal practitioners, to preserve the attorney-client privilege.[5] Katz was thus the decision that paved the road for communication privacy in the information age by significantly altering the interpretation of the Fourth Amendment and attaching a reasonable expectation of privacy to high tech wire communications.[6]

B. Cordless and Cellular Telephones

Based on the ruling of Katz,[7] as tersely put by Justice Harlan, a reasonable expectation of privacy in a communication medium is a condition precedent to investing it with Fourth Amendment protection. Federal courts have overwhelmingly held that cordless telephone communications lack such expectation. [8] Because mobile phones utilize radio waves, which are broadcast in all directions and are more susceptible to interception by a third party than traditional phones, the majority of courts have reasoned that cordless communications do not offer the requisite reasonable expectation of privacy necessary to extend constitutional protection. In fact, a Senate Report also deemed it inappropriate to make interceptions of cordless telephone communications a criminal offense because they could be easily intercepted with readily available technologies, such as an AM radio.[9]

Most federal courts have remained steadfast in that position despite the progressive growth in telecommunications technology. To mention but a few illustrative examples, in Tyler v. Berodt,[10] the Court of Appeals for the Eighth Circuit confirmed the convictions of the defendants, rejecting their assertions of privacy expectation in the conversations they made on a cordless phone. Again, in United States v. Smith, while conceding that nearly half of all American households use cordless telephones, the Fifth Circuit held that “pure radio communications” are not afforded the same Fourth Amendment protection as communications carried by land-based telephone lines because “broadcasting communications into the air by radio waves is more analogous to carrying on an oral communication in a loud [p138] voice or with a megaphone than it is to the privacy afforded by a wire.”[11] Finally, in 1995, the Fourth Circuit upheld convictions derived from intercepted conversations because the AM or FM radio signals of cordless phones “can be intercepted with relative ease. . . .” [12] The Sixth Circuit noted in the same year that “no reported decision has concluded that a cordless telephone user has a reasonable expectation of privacy in his cordless phone conversations under Title III or the Fourth Amendment.”[13]

Critics of the above position would contend that the possibility of inadvertent interception is separate from personal expectation, because most people using a mobile phone expect their conversation to be private. Lawrence Tribe of Harvard University School of Law supports this view, contending that a cautious professional criminal knows the vulnerabilities and security risks associated with mobile phones that endanger his reasonable expectation of privacy, while the average citizen does not and expects his radio wave communications to be confidential.[14] But this was precisely the argument that the defendant unsuccessfully presented in Smith. He argued “the interception of his cordless phone conversations was a search because he did not know how the cordless phone worked or that his conversations would not be private.”[15] Hardly persuaded by this line of reasoning, the Court simply stated “a subjective expectation of privacy does not, by itself, give rise to Fourth Amendment protection.”[16]

As for the legal status of cellular telephones, it is not entirely clear because no federal court has explicitly held that Fourth Amendment protection is unavailable for cellular telephone conversations. On the one hand, one could argue that they are not protected by the Fourth Amendment because they fall in the category of oral communications rather than wire communications. The technical distinction between oral and wire communications was made three decades ago in United States v. Hall, [17] in which the court of appeals found that oral communications broadcasted into the air by radio waves, as opposed to wire communications involving two regular land line telephones, failed to meet the Fourth Amendment test under Katz, because the former can be readily intercepted. But the Hall court also found that if one party to a communication used a land line telephone, then the conversation would be considered [p139] a wire communication for constitutional purposes, even if the other party was using a radiotelephone.[18]

On the other hand, there is reason to believe, in light of more recent case law, that oral cellular communications could be protected by the Fourth Amendment. In Shubert v. Metrophone, Inc.,[19] for instance, several plaintiffs sued their cellular telephone carriers for not using the available scrambling or encrypting techniques, thereby making their calls easily susceptible to interception. The Third Circuit affirmed the dismissal of the complaint because it was founded on “the mere facilitation of cellular transmissions,” rather than the improper acquisition of the contents of the plaintiffs’ communications by intentional interception. [20] Although the claim was lodged on statutory rather than constitutional grounds, the gist of the opinion is that cellular telephone users are entitled to expect privacy. Moreover, in United States v. Mathis, [21] the Eleventh Circuit (indirectly) recognized the existence of a privacy interest in cellular phone conversations based on federal law. At issue was the introduction of private conversations made by the defendant on cordless and cellular telephones that were intercepted by police with neither his consent nor prior judicial approval, which the lower court had admitted as lawful evidence in 1992. On appeal, the circuit court “agree[d] with the district court to the extent that, at the time [the defendant’s] conversations were intercepted, federal statutory law recognized no reasonable expectation of privacy on a cordless telephone.” [22] In other words, the circuit court implied that a warrant was not necessary at that time to intercept such communications but might be necessary now under new congressional amendments that were subsequently adopted. In so holding, the circuit court tacitly recognized that oral communications conducted over cellular telephones warrant legal protection.

In another 1996 case, United States v. Gaytan, the government’s evidence “consisted of tape recordings and transcripts of the defendants’ cellular telephone conversations obtained through a wiretap,”[23] as required by federal law.[24] Why would government agents have to apply for a [p140] warrant before wiretapping private cellular telephone conversations unless the user had justifiably and reasonably expected that his communications would not be subject to interception? This conclusion is further substantiated by the fact that most cellular phones, unlike their cordless counterparts, now scramble their signals and operate at a frequency range with a minimum risk of interference or interception, unless a specially adapted scanner is used.[25]

C. Fax Communications

Communications carried by faxes are similar in vein to other forms of protected wire communications in that they consist of digital signals transmitted over a traditional wired network. Facsimile transmissions, however, whether sent from computer to computer or fax machine to fax machine, do run the risks of inadvertent misdialing and misdirection due to human error. In spite of the absence of absolute security, a properly directed and received fax has the convenience of almost instant correspondence, as well as the advantage over a telephone conversation of providing the parties involved with a tangible paper record of every transaction and instance of communication.

The case law on fax communications is extremely meager on both state and federal court levels. At least one state court has found that faxes exchanged between attorneys and clients are confidential and privileged,[26] which indicates that a reasonable expectation of privacy exists in faxed documents. Although no federal court to date has so held, it can be fairly assumed that such finding is in order given that many courts have begun accepting faxed legal documents and utilizing the fax machine to correspond with attorneys.[27] Some commentators have also argued for the legitimacy of the privacy interest in fax communications because they are not easily intercepted and are relied upon perhaps more heavily than telephones in today’s business world. Jarvis and Tellam, for instance, are of the opinion that “unquestionably a fax that is not misdirected can be the subject of a claim of privilege,” and hence subject to constitutional [p141] protection because “the expectation of confidentiality in attorney-client relationships and the expectation of privacy in the context of Fourth Amendment for search and seizure purposes are largely parallel concepts.”[28]

D. Telex Transmissions

Turning to telex messages, the telecommunication predecessor of e-mail, we immediately realize that federal courts have a mixed and inconsistent record in treating them as reliable evidence, although they are transmitted over land-based telephone lines. In the case of United States v. Kim,[29] the Court of Appeals for the District of Columbia refused to admit into evidence a telex message offered by the defense because, among other reasons, its “method or circumstances of preparation indicate lack of trustworthiness” because of its inconsistency with other supporting evidence.[30] Since the Kim court focused more on the context of the pertinent transactions and less on the fact that the evidence consisted in telexes, it is not clear whether the Kim court excluded the evidence because of circumstances unique to the case or the status of telexes as records. The First Circuit also refused to admit telexes into evidence on “questionable trustworthiness” grounds,[31] as did the Second Circuit, which noted “telexes of this sort are not business records and contain many inaccuracies.”[32] But again, it is not entirely clear whether their findings were based on the features of the cases or ambivalence toward the nature of telexes, and this issue has not reappeared before the courts.

At any rate, other federal courts have admitted telex messages into evidence. [33] Of particular interest to us is United States v. Gregg,[34] in which the Court of Appeals for the Eighth Circuit admitted court-authorized interceptions of telex communications introduced by the government and affirmed the lower court’s decision. When the defendants argued in the district court that the telex interceptions did not meet the requirements of [p142] the Fourth Amendment, the district court did not sweep aside their contention as irrelevant, but went on to examine whether probable cause existed for the telex warrants, whether the warrants were based on the supporting affidavits, and whether they met the particularity requirements. [35] In finding that the warrants complied in all respects with Fourth Amendment requirements, both courts in effect confirmed that a legitimate expectation of privacy exists in telex communications.

E. E-Mail Messaging

We finally get to electronic mail, one of the latest developments in popular communication technology. A revolution in telecommunications, e-mail allows computer users to efficiently send messages and data files across the country (or even the globe) almost instantly, and to keep an accurate and permanent record of this exchange for later reference on both the sender’s and recipient’s computers. Sending information by e-mail has the additional advantages over telephone and traditional mail of never getting a busy signal and free, speedy delivery. It is also more practical than faxes because it does not require that a recipient be present at the receiving end to pick it up and it can be sent at once to more than one destination and retrieved from more than one location.[36]

The convenience, speed, cost-effectiveness, and wide accessibility of e-mail have made it a tremendous success, as evident from the volume of e-mails transmitted by businesses in the United States, which is estimated at 3.5 billion per day. [37] The use of e-mail and the Internet has become an indispensable part of the modern business world, holding an estimated sixty percent of the average company’s most vital business data. [38] A survey conducted in the early 1990s revealed that e-mail was being used in some capacity by seventy-five percent of all large companies in the United States.[39] A more recent study found that, based on past trends and growth [p143] prospects, the number of employees corresponding via e-mail is expected to increase by twenty percent every year.[40]

Like postal mail, every e-mail message is directed to a unique, personally identifiable address and delivered to a password-protected electronic mailbox, where unread messages await the user to access the mailbox and open them. The Internet, which is the existing version of the Information Superhighway, serves as the primary medium in which the global e-mail network operates.[41] It consists of a worldwide network of computers that communicate with each other using hard telephone lines or satellites. Before it reaches the recipient’s mailbox, a sent Internet e-mail message travels through several intermediate computers and leased lines, known as hosts and routers, which are often owned and operated by third parties. [42] A copy is made of every transmitted e-mail message as it passes through these intermediary hosts or routers, which may be lawfully used or disclosed by router employees to the extent necessary to protect its rights or render the service.[43] This so-called store-and-forward technology [p144] is necessary to ensure message arrival, since electronic mail must travel through several computers before it reaches its intended recipient, but it raises serious privacy concerns on the part of users.[44]

The development of e-mail constitutes yet another challenge to government because e-mail communications may be used for illicit or illegitimate purposes. Detection and prevention of these activities entails invading the privacy of users by accessing and sometimes monitoring their e-mails. At the societal level, government may have to gain access to private e-mail for investigative purposes to gather after-the-fact evidence of criminal behavior. In other instances, it may have to intercept e-mail traffic in real time to prevent the consummation of an illegal scheme. [45] But the public value of privacy in communication would be unreasonably compromised if government were allowed to conduct indiscriminate searches of private e-mails.

At the organizational level, Reg Whitaker, a noted authority on information technology, argues that all employers have a valid interest in verifying that their e-mail systems are not being used for unethical purposes and that their employees are not engaged in criminal activity.[46] He refers to the Iran-Contra affair as a good example of how public interest could be served by accessing the e-mail correspondence of government employees, because Oliver North, among other conspirators, was identified by checking archived e-mails. [47] On the other hand, Amitai Etzioni, a communitarian sociologist, argues that the lack of e-mail privacy adversely affects organizational culture and undermines the sense of community in the work place. [48] The concessions that need to be made for law enforcement purposes in order to balance the conflicting interests at stake will be discussed later in this Article. [p145]

III. The Relevant Case Law

The American Management Association reports that forty percent of major American corporations monitor their employees electronically in the normal scope of business, using a variety of methods that range from counting their computer keystrokes to reading their electronic mail, and from listening to their telephone conversations to observing them by overhead closed-circuit television cameras. [49] While millions of Americans work under the watchful and electronic eye of their bosses, they cannot invoke the Fourth Amendment whose strictures are directed against the officers of the state to the exclusion of private citizens. Unlike public organizations, privately owned businesses are only bound by statutory law and congressional legislation, such as Title III of the Omnibus Crime Control and Safe Streets Act of 1968, the Privacy Act of 1974, and the Electronic Communications Privacy Act of 1986 that Congress passed as an amendment to Title III, with the intention of extending its coverage to new types of digital communications technologies including e-mail, although it remains largely untested in that area. [50] Admittedly, monitoring technologies have intensified employee privacy concerns in the public and private sector alike, but given the nature of this study, the latter is not part of our inquiry which focuses on Fourth and Fifth Amendment laws that apply only in governmental work settings.

Justice Clark once lamented “the law, though jealous of individual privacy, has not kept pace with these advances in scientific knowledge.” [51] Because the U.S. Supreme Court has not yet undertaken to reconsider the balance between the government’s interests and the individual’s right to privacy, or to redefine the contours of Fourth Amendment protection in light of this new technology, the legal boundaries of e-mail technology remain somewhat hazy. To draw our conclusions, then, we will have to rely on the opinions of lower courts, and the possible legal and technical parallels between e-mail and the other forms of communication discussed above.

E-mail presents the judiciary and lawmakers with a difficult case because it falls somewhere between a telephone communication and a [p146] written correspondence via postal mail.[52] It resembles a telephone call because it consists of intangible electronic signals traveling through wire systems, and a first class letter in that the data in the transmission contains a nonvocal textual message. But while e-mail is a cross between both, it is afforded the privacy protection of neither against government interception and acquisition.

Based on Katz, police must obtain a search warrant from a judge in order to wiretap a telephone line or listen to a private conversation, whether made from the home or a public telephone booth. Further, the government must establish to the judge authorizing the wiretap that normal procedures have been tried but failed to serve its investigative interest and that the communications to be intercepted are linked to the commission of an offense. As for searches of first class mail, they too can only be done pursuant to a court order but are even more narrowly drawn than telephone wiretaps. [53] While probable cause to engage in electronic eavesdropping is particularized to a telephone user and subjects all his calls to government surveillance, probable cause to open first class mail is not particularized to a sender but to pieces of mail directed to specific recipients and hence less general. [54] Thus, both modes of communication have full Fourth Amendment protection from unreasonable searches and seizures, whereas e-mails, as will be seen from the cases below, have not been afforded the same level of protection.

With the ubiquity of e-mail in the work place, e-mail evidence has lately come to play a key role in litigation in general and employment lawsuits in particular. More courts are now admitting e-mails as relevant and competent evidence. For instance, in Aviles v. McKenzie,[55] a wrongful termination and employment discrimination case, the Second Circuit accepted the e-mail presented by the plaintiff to buttress his claims as viable evidence. Again, in Strauss v. Microsoft Corp.,[56] e-mails received by the plaintiff from her supervisor were used as the principal piece of [p147] evidence at trial. In fact, the district court noted that these e-mails were sufficiently persuasive as to “lead a reasonable jury to conclude that . . . [Microsoft] failed to promote Strauss as a result of gender discrimination.”[57]

We now turn to e-mail cases involving the application of the Fourth Amendment. The first major decision in the area of e-mail search law was United States v. Maxwell,[58] a case of criminal misconduct through transmissions and communications using e-mail accounts provided by a private on-line service. James A. Maxwell, an America On-Line (AOL) subscriber, had the exclusive use of four separate screen names or identities. [59] The FBI received a tip that Maxwell was using his AOL e-mail account to transmit and receive, in violation of federal law, visual images of minors engaged in sexually explicit conduct. Following discussions with the informant and AOL officials, the FBI applied for and obtained a search warrant permitting the seizure of information pertaining to suspected subscribers in nine AOL computers. The search yielded visual transmissions of child pornography bearing, among other AOL user names for which they had a warrant, one of Maxwell’s screen names. Included within the files that the FBI seized were private e-mail communications made by the defendant under a different screen name to another Air Force member, in which he lucidly discussed his sexual preferences and desires. These communications were used as the basis for an additional charge of communicating indecent language.

The Air Force Criminal Court of Appeals found that the search and seizure of the defendant’s e-mail communications were controlled by the unreasonableness clause of the Fourth Amendment because he “definitely maintained an objective expectation of privacy in any email transmissions he made so long as they were stored in the America On-Line computers.”[60] Maxwell “clearly” had an objective expectation of privacy in the stored messages because “he alone could retrieve [them] through the use of his own assigned password.”[61] Likewise, the defendant also had an objective expectation of privacy in the e-mails transmitted to other AOL subscribers because they all had individually assigned passwords. The Maxwell court [p148] also noted that computer e-mails, unlike cordless phone transmissions or telephone calls, could not be “received by anyone other than the intended recipients.”[62]

Turning to the second question of probable cause, the court of appeals determined that probable cause was required to seize information under each separate e-mail account. The government established probable cause in so far as identifying the defendant as the user of one screen name, and hence, to seize the e-mail transmissions made only under that name, but not the other three screen names that were readily furnished to the FBI by AOL. Bureau agents investigating the matter were unaware of the existence, much less the criminal behavior, of the other identities; it was AOL personnel who made the connections between the multiple identities for the FBI. This is the reason why none of the other three names appeared on the search warrant.[63] Moreover, the court of appeals likened the search of electronic mailboxes to mailboxes at the post office, where one person may rent several boxes for different purposes. That probable cause exists to search one particular box does not mean that the government is authorized to search all the other boxes possessed by the same person. Accordingly, the court of appeals concluded that the seizure of e-mails for those screen names that were initially unknown to the FBI and unstated on the warrant was constitutionally impermissible, because there was no probable cause to believe that a search of the files stored under those names would reveal evidence of a crime.[64]

In June 1996, the Court of Appeals for the Armed Forces granted review to the Maxwell case.[65] The reviewing court approved most of the findings of the appellate court below, but added several points that are very instructive. In answering the constitutional questions before it, the reviewing court examined a number of issues that could be applicable to other e-mail search cases, of which we consider the ones most relevant to our inquiry. First, the technology used to communicate via e-mail. In this respect, the reviewing court found that technology is “extraordinarily analogous to a telephone conversation,” and hence “protected by the Fourth Amendment if there is a reasonable expectation of privacy.” [66] Second, with regard to the relationship between the defendant and AOL in terms of contractual obligations, the court of appeals found that AOL’s agreement with Maxwell ensured his privacy. Finally, the court of appeals [p149] considered the “tenor and content of the email conversations” between the defendant and the other AOL subscriber. [67] The character of these communications revealed that the correspondents had a reasonable expectation that their conversations were private.[68] It is noteworthy that the court of appeals specifically stated that the possibility that “an unauthorized ‘hacker’ might intercept an email message does not diminish the legitimate expectation of privacy in any way.”[69]

Additionally, the court of appeals noted that AOL differed from other systems in that its e-mail messages “are afforded more privacy than similar messages on the Internet,” because they are privately stored for retrieval on the company’s privately-owned computer bank, and are not read by or disclosed to anyone — including AOL executives — other than the authorized users.[70] It follows that AOL users not only have a reasonable expectation of privacy in their e-mail messages, but also in those kept in AOL’s computers. It also follows that the police need obtain a warrant not only to conduct real time surveillance of e-mail communications, but also to search old files archived within the networks of service providers.

One analyst agrees with the court’s ruling, maintaining that AOL, along with similar services such as Prodigy and Compuserve, should be classified as semi-public hosts that act as gateways to the Internet and Internet e-mail, because they are relatively insulated from the worldwide cyberspace.[71] This places AOL in a middle category between Local Area Networks (LAN) that utilize internal lines and open systems that utilize outside lines. In legal terms, there is a higher expectation of privacy in e-mails of such private on-line networks as AOL that afford their users considerable privacy than in other e-mail accounts whose e-mails stream across the open Internet. It goes without saying, however, that this expectation hardly exists in e-mails transmitted over the exact same network when directed to strangers (e.g., chat room participants or Usenet newsgroup members), because the sender of such e-mail “runs the risk that he is sending the message to an undercover agent,” as one federal judge simply put it in United States v. Charbonneau.[72]

Bohach v. City of Reno[73] is another Fourth Amendment case involving access to electronic messages stored on computers. But unlike Maxwell that shed light on the nature of expectations of privacy in e-mail in the [p150] context of a private provider of electronic communications, Bohach revolves around expectations of privacy in messages created and recorded on a government-owned and government-operated computer system. The computerized paging system (Alphapage), which was designed to free up the regular telephone lines of the Reno Police Department, allowed the user to compose and send brief alphanumeric or voice messages by keyboard or telephone to the recipient officer’s pager over a LAN system. All the transmitted alphanumeric messages, which the federal court described as “essentially electronic mail,” were stored on the department’s routing computer.[74] Bohach and Catalano, both members of the police force, came under an internal affairs investigation because of the content of the messages they sent to one another. They brought action to stop the investigation and bar the disclosure of their electronic messages, claiming that their storage on and retrieval from the department’s computer network was a violation of their constitutional right to privacy.

The district court agreed that the officers had a subjective expectation of privacy simply because they would not have sent the sorts of messages that triggered an investigation had they thought otherwise.[75] However, it concluded that it was not the kind of expectation that was objectively reasonable based on a number of factors. First, there was no interception or wiretapping of communications; the storage of sent information was part of how the system worked. [76] Second, the system was accessible to virtually anyone with access to or working knowledge of it, and required no special password or clearance to use it.[77] Third, the primary purpose of the system was to allow work-related and not private communications between police personnel.[78] Fourth, the fact that certain types of messages — such as comments on the department’s policies — were banned from the system suggested that users should anticipate less privacy. [79] Finally, the computer service provider, that is the City of Reno, may access communications in electronic storage as it pleased pursuant to the city ordinances.[80]

Despite the promise of the Maxwell precedent, the Bohach court found that the affected parties had no reasonable expectation of privacy in the communications they exchanged over a LAN computer system. There are several key differences between the two cases that led to the different conclusions. The first distinction between them is that the defendant in [p151] Maxwell purchased all of his computer hardware, software, and Internet service subscriptions with his personal funds, while the plaintiffs in Bohach used the police department’s terminal, computers, software, and pagers. Another important distinction, which is a product of the first, is the nature of the uses to which each system was intended. Whereas AOL had no connection whatsoever with the officer’s official duties and was expressly not subject to monitoring, Alphapage was installed to facilitate police work and allow for information flow between police personnel in a timely fashion, and hence, implicitly subject to monitoring. In fact, the Bohach court stated that it would have reached the same conclusions even if an interception by the provider had actually occurred, [81] which means that Bohach and his fellow-officers were not entitled to harbor any privacy expectations in messages transmitted over the department’s computer system.

Based on this ruling, one could argue that Maxwell should not have had a legitimate expectation because his correspondent, who was a fellow officer, could have forwarded his pornographic messages to the authorities. To rebut this argument, we need to determine the point at which the Fourth Amendment ceases to shield the electronic messages from government access. In Bohach, Fourth Amendment protection did not exist in the first place because of the real possibility, if not the implicit consent, that police may monitor all communications between on-duty officers made over the department’s computer network. In Maxwell, by contrast, Fourth Amendment protection remains in full effect until one party betrays the confidence of the other. Only then is the constitutional protective guard lifted because “the Fourth Amendment does not protect a wrongdoer’s misplaced belief that a person to whom he voluntarily confides his wrongdoing will not reveal it.”[82]

United States v. Monroe[83] posed more or less the same legal question that Bohach answered, that is, whether a government officer has a reasonable expectation of privacy in an e-mail account owned by the state and is subject to monitoring. In Monroe, however, users gained access to the system and the Internet through a personal log-on name and a private password. Although e-mail was essentially reserved for official business because all accounts resided on a computer owned by the U.S. Air Force, on-base officers were allowed to use their e-mail to send and receive personal messages from friends and family. Excessively large messages that accumulated in the directory were automatically deleted after seventy-[p152]two hours, but software errors sometimes prevented this automatic deletion. As a result of an occasional system error, the two sergeants charged with the network administration found fifty-nine messages that had been stuck in the system for more than three days. Without opening any of the lodged messages, they discerned that they could be pornographic in nature from their sexually oriented names. Many of the suspected messages were addressed to a user known as “monroer.” [84] The sergeants’ suspicions were confirmed when they opened some of the messages and discovered they contained sexually explicit photographs of women. This user name was traced back to the e-mail account of the defendant, Robert J. Monroe, which the administrators accessed to find out if he had requested those images or was just the victim of a prank. In one of the messages sent to the originator of the fifty-nine e-mails, they read a reminder from Monroe to “send the file.”[85]

A special investigation ensued, which culminated in the issuance of a warrant authorizing the search of Monroe’s dormitory room and the seizure of all computer-related data media suspected to contain pornography. The search produced a large number of graphic and textual computer files that were obscene as legally defined, some of which contained child pornography. Monroe unsuccessfully moved to suppress all evidence, challenging the initial viewing of his stuck e-mails and the subsequent invasion of his quarters as illegal searches, wanting in probable cause.

Affirming the decision of the Air Force Court of Criminal Appeals, the Court of Appeals for the Armed Forces held that Monroe had no reasonable expectation of privacy, not only in the e-mails that got lodged in the network directory, but also in the ones stored in his personal e-mail box. The Monroe court noted that this case differed from Maxwell in several respects that warranted a different conclusion. First and foremost, “AOL contractually agreed not to read or disclose subscribers’ email to anyone other than authorized users,” whereas anyone logging into the Air Force system received a specific banner message stating that “Users logging onto this system consent to monitoring by the HOSTADM.”[86] Second, the system here was owned by the government, which means that no privacy interest could be asserted by consenting users, at least against personnel officially responsible for maintaining the network.[87] By the same [p153] token, the two sergeants in charge of system administration committed no illegality when they tried to ascertain the nature of the messages clogging the system.

Monroe confirms that government employees using government computers have no expectation of privacy with regard to their e-mails and any information stored on government-provided resources. The Fourth Amendment then does not protect a government employee until law enforcement authorities decide to extend the scope of the search beyond the information seized from the central computer system or the personal workstation. This means that the warrant requirements do not apply to either the interception of communications or the access of stored files, [88] which makes the Monroe ruling the very opposite of Maxwell.

IV. Constitutional Analysis and Conclusions

As we have seen, relatively little case law exists to date that has explored the concept of the Fourth Amendment right to privacy in the context of e-mail. But while the foregoing cases may not have fully addressed the matter, they provide us with adequate tools to develop a constitutional standard for protecting the contents of e-mail messages. As noted by the courts, similarities exist among various interpersonal and interorganizational methods of communication, which could help adjust the traditional legal boundaries to encompass the new e-mail technology that has rapidly come into widespread use at home and at work.

Although the U.S. Supreme Court has not heard a e-mail search case yet, commentators anticipate that the current interpretation of Katz would carry over to e-mail communications. The Court has frequently taken the position under Katz that the mere possibility of exposure to the public eye diminishes and sometimes obviates one’s privacy expectation. Based on this premise, the Court may find that e-mail does not warrant Fourth Amendment protection because, much like cordless telephone calls, people perceive e-mails as susceptible to interception, not to mention the increasing risk of stolen information in networked environments. Simply stated, the Court may use the lack of security during the transmission of messages and the dramatic rise in hacking and computer crime as reasons to withhold Fourth Amendment protection altogether from e-mail messaging.

[p154] Though consistent with the Court’s holdings in the area of privacy, this rationale bases the reasonableness of the users’ privacy expectations almost exclusively on the drawbacks in the existing technology or on the limitations of the informational infrastructure. But even if we follow this line of reasoning to its logical conclusion, e-mail should still be constitutionally protected as long as privacy technology can be sufficiently enhanced to eliminate or at least to reduce the security risks substantially.[89] One alternative, encryption technology, makes e-mail transmissions highly secure because it scrambles the message contents until the message reaches the intended recipient. Even if someone copies or captures an encrypted message en route, only the holder of the encryption key may read the message. Few, if any, would argue the unreasonableness of a privacy expectation in this kind of Internet communication because strong encryption makes e-mail practically impossible to decrypt and virtually pointless to intercept.

Since encryption software has solved the security problem, then encrypted e-mails should not have a lower standard of Fourth Amendment protection than paper mail, even under the Court’s narrow application of the Katz test. [90] Orin Kerr reasons that if by locking a container a person creates a reasonable expectation of privacy in its contents, then an encrypted message should receive the same Fourth Amendment protections against the warrantless seizure or decryption by law enforcement officials. [91] If the concept of physical intrusion no longer controls Fourth Amendment inquiry, it makes no sense to extend constitutional protection to a locked container or a sealed package in transit, [92] and then withhold the protection from a locked electronic message in cyberspace traffic.

Further, in emergency conditions, or where probable cause exists, the Court could treat e-mail searches as telephone wiretaps, thereby allowing law enforcement agents to read all the messages on a suspect’s account in the same manner law enforcement can listen to every conversation passing through the phone line. Courts may deem an overly broad search acceptable, if not inevitable, for telephone communications because the suspect destroys the evidence of criminality the moment the suspect hangs up the phone, leaving no tangible trail of the verbal exchange for the police to sift through. [93] Police must intercept and record all incoming and [p155] outgoing calls to preserve such volatile evidence. This rationale, however, does not hold in the case of e-mails where the evidence remains available for search and seizure long after termination of the communication, whether on the suspect’s or the service provider’s computer. Therefore, an overly broad warrant authorizing the police to monitor all sent and received e-mails repulses the Fourth Amendment as much as the general writs of colonial America the Framers expressly sought to ban. A special software program of the sort that AOL officials used in Maxwell to locate specific messages could be used by government to automatically extract from stored or intercepted messages the information that the police seek without allowing excessive invasion of privacy or leaving too much to the discretion of the officers executing the warrant.

The recent use of pocketsize wireless pagers to receive e-mail messages raises an interesting Fourth Amendment question as to whether government may explore the contents of one’s pager as part of a warrantless search incident to an arrest. United States v. Reyes[94] answered this question in the affirmative. Federal agents seized a pager from Reyes in the course of a lawful arrest, but searched its memory without obtaining a warrant. The Reyes Court upheld the warrantless search of the pager under Chimel v. California [95] as a search of a container incident to a valid arrest. Megan Bertron contends that the warrant exception of Chimel is inapplicable in this search situation because a pager, unlike a closed box, cannot be used to hide a dangerous instrument or a deadly weapon. [96] Moreover, if the pager poses no threat to the safety of the arresting officer, then its search requires another warrant based on a separate determination of probable cause.[97]

Also, the search of Reyes’ pager did not protect against the concealing or destroying of evidence because police had no reason to believe that the pager contained information valuable to them in the first place. Undertaking an immediate search in absence of knowledge that the pager constituted destructible evidence makes the search a “blind fishing expedition”[98] by which the arresting officers hoped to discover rather than preserve incriminating evidence. In fact, previous rulings point out that police should wait for the evaluation of a neutral magistrate rather than conduct a prompt, warrantless search absent adequate justification because “evidence may not be introduced if it was discovered by means of a seizure [p156] and search which were not reasonably related in scope to the justification for their initiation.”[99] Therefore, unless the police have reason to believe that a pager “contained evidence which would lose its value,”[100] the police should not search it incident to a lawful arrest.

Courts will also need to address the important question of whether government may gain possession of private e-mails by a subpoena. The police could serve a subpoena on the service provider to produce all e-mails sent or received by one or more subscribers stored in its computers, or on the individual subscribers themselves, commanding them to produce their own e-mails. Drawing on the courts’ treatment of other means of communication, nothing in the Fourth or Fifth Amendments prohibits government from procuring e-mails by the first means. For instance, courts could extend the rationale of Newfield v. Ryan,[101] a telegraph case, and draw this inference to e-mail. There the government served a telegraph company with a subpoena requiring the production of “any and all telegrams or copies thereof in [their] custody or control”[102] sent or received within a limited period of time by certain clients pertaining to specific transactions. The affected clients brought an action, attacking the subpoena under the Fourth and Fifth Amendments, and complaining that the subpoena compelled them to testify against themselves. The circuit court disagreed, holding that, because the circuit court served the subpoena on the telegraph companies, the plaintiffs have “no standing whatever” to invoke the Fifth Amendment because they “are being called upon neither to produce evidence, nor to testify against themselves, they are not being called upon at all.” [103] Nor do the plaintiffs have standing to invoke the Fourth Amendment because they are not “being subjected to a search and seizure, reasonable or unreasonable, as to their persons, or their properties.”[104]

If courts can subpoena telegrams, which bear the closest resemblance to e-mails, then courts can subpoena e-mails too. The subpoena procedure in practice requires less than the warrant procedure. Rather than the probable cause requirement for warrants, to obtain material through a subpoena the government need only present a court with “specific and articulable facts showing that there are reasonable grounds” to demonstrate [p157] that the communications sought relate to an ongoing criminal investigation.[105]

United States v. Barr[106] also substantiates this position. In this case, the district court served a subpoena on a private company (Affiliated) that Barr, a suspect for engaging in narcotics-related activities, employed to receive his mail and telephone messages. The district court served a subpoena on Affiliated requesting the production of mail addressed to “Larry Freeman,” an alleged alias for the defendant. [107] The district court rejected Barr’s claim that the subpoena violated his right against self-incrimination and that it used the subpoena to circumvent the warrant requirements since “the agents who served the subpoena did not coerce compliance by force or threats or overstep their legal authority” and “the subpoena was not overly broad.”[108] Moreover, the burdens imposed on a suspect by the subpoena process likely do not offend the Fourth or Fifth Amendment unless a subpoena lacks specificity, or “when government agents improperly impinge on the defendant’s right to contest the subpoena’s validity or a court’s authority to quash, alter or enforce it.” [109] Again, these rules equally apply to a subpoena compelling the production of e-mails from a third party.

Courts could reach a different conclusion, however, regarding the procurement of e-mails from the subscribers themselves by subpoena. E-mail distinguishes itself from postal mail and other forms of communication in that the drafter may (optionally) keep a copy of every sent message on the hard drive of the computer. This carbon copy, kept for personal reference, contains private communications whose introduction at trial could turn the sender into an involuntary witness against himself and therefore give rise to constitutional concerns, since a court serves the subpoena on him and the words sought are his. A different situation exists when obtaining the same message from the receiver or the service provider because of the personal nature of the Fifth Amendment which only the holder may assert to bar the production of documentary evidence.[110]

[p158] United States v. Hubbell, [111] supports the above conclusion. There the U.S. Supreme Court held that a broadly-worded subpoena calling for the production of diaries and personal electronic mail messages, among other personal records, violated the Fifth Amendment. Precisely stated, a subpoena runs afoul of the Fifth Amendment if government lacks “any prior knowledge of either the existence or the whereabouts” of the subpoenaed evidence.[112] The act of producing such confidential documents, that have a questionable existence, has a testimonial aspect in itself because it implicates the self-incrimination clause forcing the defendant to acknowledge the “existence, possession, and authenticity” of the sought documents.[113] Therefore arguing under Hubbell, the government may not serve an e-mail user with a subpoena compelling the production of stored e-mail messages since not all e-mail users always keep a copy of their sent e-mails on their personal machines.

Moreover, the testimonial and communicative nature of e-mail affords it further protection from government access. To define the ambit of the privilege against self-incrimination, the Hubbell Court clarified what constitutes a compelled testimony by noting two important points applicable to e-mail. First, it noted that “[t]he word ‘witness’ in the constitutional text limits the relevant category of compelled incriminating communications to those that are ‘testimonial’ in character,” [114] and second, that “there is a significant difference between the use of compulsion to extort communications from a defendant and compelling a person to engage in conduct that may be incriminating.” [115] Because e-mail stored on a person’s hard drive has both testimonial and communicative characteristics, then its introduction at trial over the objection of its owner raises both Fourth and Fifth Amendment concerns despite the fact the writer expressly uses it to convey information to another party. This constitutes the quintessential evidence that the Fifth Amendment protects a person from furnishing against himself. The Court may well accept this from its broad statement that the “‘right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and [p159] seizures,’ would not be violated, under any ordinary construction of language, by compelling obedience to a subpoena.”[116] This argument obviously does not equally apply to personal e-mails of public employees stored on government-owned computers who must restrict their use to work-related purposes.

One final question remains: should the courts allow the government to engage in real time monitoring of all e-mail correspondence over its systems, or to freely access any employee e-mail account on the networks that it owns? Courts should not answer the question whether courts should give the government blanket authority to conduct general searches of its employee accounts, but rather, whether the government may deprive public servants of all privacy interests in their e-mail. [117] This forces courts to make a public policy decision as much as a constitutional law decision. Judges clearly involve themselves in policymaking through the establishment of authoritative rules and in determining the scope of their application. Besides, their decisions directly influence how government agents will act and use their official authority and power.[118]

I believe probable cause and the full search warrant requirements should preserve only the rights of private citizens using semi-public carriers such as AOL, which contractually ensure the privacy of their users as already decided in Maxwell, as well as to those using privately-owned Internet services that utilize encryption technology. As for governmental work settings, I propose that the judiciary ought to create different classifications of public organizations entitling employees to different levels of Fourth Amendment protection in their use of electronic communications based on their organization’s nature or the sensitivity of its operations. Employees of the FBI, CIA, armed forces, and various law enforcement agencies, for instance, would have minimized privacy rights because of the crucial importance of their organizations to national security and public safety. This category of government organization would equate in the informational sense with the extra hazardous and pervasively [p160] regulated industries subject to warrantless periodic inspections to ensure strict compliance with health and safety standards. The surveillance of computer information systems and e-mail accounts of employees working under special demands of discipline and duty analogize to authorized inspections, which permit limited intrusions for defined purposes when a special need exists. [119] This proposal equates with the courts’ rulings in Bohach and Monroe.

Courts could create another classification for e-mail communications between employees working for public organizations not charged with intelligence or order-maintenance functions, which would subject them to the lower constitutional standard of reasonable suspicion as a balance between the more rigorous probable cause standard and the absence of all protection.[120] Under this scheme, government would have to have a reasonable suspicion in order to access the e-mail accounts of these employees, or at least present to a court an after-the-fact showing of reasonable suspicion as a precondition to admitting any e-mails into evidence. Agencies could conduct a full-fledged search later if the situation unfolds and probable cause is found. This actually happened in Monroe, where the reasonable suspicion of the sergeants gave rise to probable cause, which, in turn, prompted a legal search. But the Monroe Court declined to dwell on these technical limitations because authorities could have still performed an initial full-fledged search of all the defendant’s e-mails absent any suspicion of wrongdoing. This sliding-scale approach has the double advantage of deterring the improper use of e-mail on the part of public employees and preventing needlessly invasive behavior on the part of the state. The adoption of this alternative standard entails a case-by-case analysis because in the absence of a bright line rule, “the evaluation must be made from the facts of each particular case.”[121]

In light of the vital role the Internet and e-mail have come to play in our society, a reinterpretation of the Fourth Amendment and perhaps a whole new subfield of search law will develop. The keys to this new legal world ultimately lie in the hands of the U.S. Supreme Court, which must strike the balance between promoting technology and protecting society from the same. In developing this new body of law, however, the courts must [p161] remember that liberty, as a core value, has brought this country thus far and continues to propel it forward.