IN THE

UNITED STATES COURT OF APPEALS

FOR THE DISTRICT OF COLUMBIA CIRCUIT

Nos. 99-1442, 99-1466, 99-1475, 99-1523

UNITED STATES TELECOM ASSOCIATION,

CELLULAR TELECOMMUNICATIONS INDUSTRY ASSOCIATION,

CENTER FOR DEMOCRACY AND TECHNOLOGY, et al.,

Petitioners,

v.

FEDERAL COMMUNICATIONS COMMISSION

and UNITED STATES OF AMERICA,

Respondents.

ON PETITION FOR REVIEW OF AN ORDER OF

THE FEDERAL COMMUNICATIONS COMMISSION

REPLY BRIEF OF PETITIONERS UNITED STATES TELECOM ASSOCIATION,

CELLULAR TELECOMMUNICATIONS INDUSTRY ASSOCIATION, AND

CENTER FOR DEMOCRACY AND TECHNOLOGY

TABLE OF AUTHORITIES

SUMMARY OF ARGUMENT

ARGUMENT

I. CALEA REQUIRES ONLY THE CAPABILITY TO ISOLATE AND DELIVER CALL CONTENT AND CALL-IDENTIFYING INFORMATION COEXTENSIVE WITH THE SURVEILLANCE AUTHORITY DEFINED BY TITLE III AND THE PEN REGISTER STATUTE

A. CALEA's Capability Requirements Are Limited to the Information Specified in the Two Principal Statutes Defining Law Enforcement's Electronic Surveillance Authority

B. Contrary to the Government's Assertions, the Order Defines CALEA's Minimum Requirements and Will Result in Intrusions on Privacy That Will Not Be Subject to Later Judicial Review

II. LOCATION INFORMATION AND THE INFORMATION PROVIDED BY THE FOUR ADDED CAPABILITIES ARE NOT "CALL-IDENTIFYING INFORMATION" REQUIRED BY CALEA

A. The Government Cannot Overcome Petitioners' Demonstration That "Call- Identifying Information" Is Limited to the Numbers Used To Route Calls

B. The Government Offers No Reasoned Statutory Basis for Concluding That Location Information and the Four Added Capabilities Are "Call-Identifying Information"

1. Location Information

2. Dialed Digit Extraction

3. Hold/join/drop on Conference Calls

4. Subject-Initiated Dialing and Signaling Information

5. In-band/out-of-band Signaling

III. THE FCC DID NOT MEET EVEN MINIMAL STANDARDS OF REASONED DECISIONMAKING IN CONCLUDING THAT THE CAPABILITIES IMPOSED BY ITS ORDER ARE "REASONABLY AVAILABLE" AND SATISFY THE APPLICABLE STATUTORY CRITERIA

IV. THE ORDER IMPERMISSIBLY REQUIRES THE DELIVERY OF PACKET-MODE COMMUNICATIONS INCLUDING CALL CONTENT

CONCLUSION

Brown v. Waddell, 50 F.3d 285 (4th Cir. 1995)

Edward J. DeBartolo Corporation v. Florida Gulf Coast Building and Construction Trades Council, 485 U.S. 568 (1988)

Halverson v. Slater, 129 F. 3d 180 (D.C. Cir. 1997)

Marshall v. Barlow's, Inc., 436 U.S. 307 (1978)

Smith v. Maryland, 442 U.S. 735 (1979)

United States v. New York Telephone Company, 434 U.S. 159 (1977)

Title III, Omnibus Crime Control and Safe Streets Act of 1968, 18 U.S.C. § 2510 et seq.

Electronic Communications Privacy Act of 1986, 18 U.S.C. § 3121 et seq.

Communications Assistance for Law Enforcement Act, 47 U.S.C. § 1001 et seq

47 U.S.C. § 201(b)

Digital Telephony and Law Enforcement Access to Advanced Telecommunications Technologies and Services: Joint Hearings on H.R. 4922 and S. 2375, 103d Cong. 6 (1994)

H.R. Rep. No. 103-827 (1994), reprinted in 1994 U.S.C.C.A.N. 3489

In the Matter of Communications Assistance for Law Enforcement Act, Report and Order, 14 FCC Rcd 4151 (1999) 5

In the Matter of Communications Assistance for Law Enforcement Act, Third Report and Order, 14 FCC Rcd 16794 (1999) passim

The government views CALEA through the prism of law enforcement and thus sees in CALEA only the opportunities for law enforcement and none of its limitations. The government therefore does not analyze the statutory language and its structure; it offers instead only the same ipse dixit pronouncements originally embraced by the FCC. And it overlooks or ignores the provisions of CALEA that require carriers to comply with law enforcement needs only "in a manner that protects the privacy and security of communications and call-identifying information not authorized to be intercepted." 47 U.S.C. § 1002(a)(4)(A) (emphasis added).

The government's one-dimensional approach allows it to rationalize the Order's intrusive regime on the fanciful notions that the ex parte surveillance orders that it presents to judges will invariably be narrowly drafted and that law enforcement agencies should be trusted to abide by self-imposed minimization efforts in processing the information that the Order will allow to fall into their hands, but to which they are not entitled. But Congress made clear its intention that unnecessary and unduly expansive collections of information not occur in the first place.

Nor does the government's infinitely elastic rationale for imposing massive costs on carriers — that the costs are "not so exorbitant" — constitute reasoned decisionmaking. Ultimately, the government's arguments utterly fail to recognize Congress's insistence that CALEA be implemented by according equal dignity to fundamental national priorities of privacy, minimizing costs to carriers and consumers, and continued technological innovation, along with law enforcement interests.

ARGUMENT

CALEA Requires Only the Capability to Isolate and Deliver Call Content and Call-Identifying Information COEXTENSIVE with the Surveillance Authority Defined by Title III and the Pen Register Statute.

The government contends that this Court should entirely ignore the underlying electronic surveillance laws and focus only on carriers' "technical capability" to provide information to law enforcement, because law enforcement purportedly will use surveillance information "when — and only when — [it] has the necessary legal authority." (DOJ Br. at 1-4.) This approach conflicts with CALEA's plain language and structure, ignores Congress's explicit mandate to protect the privacy of communications, wrenches CALEA from the law enforcement surveillance statutes on which it rests, and defies common sense.

1. CALEA's Capability Requirements Are Limited to the Information Specified in the Two Principal Statutes Defining Law Enforcement's Electronic Surveillance Authority.

The government wrongly asserts that CALEA's requirements can be discerned without reference to the two principal statutes limiting law enforcement's authority to conduct electronic surveillance. But CALEA requires carriers to invest billions of dollars in supporting electronic surveillance capabilities, and Congress plainly would not have required industry to invest immense resources on developing capabilities that law enforcement may not lawfully use. To the contrary, Congress carefully limited industry's obligations to the "narrowly focused" capabilities required to support the two principal electronic surveillance statutes, Title III and the pen register and trap-and-trace sections of ECPA. H.R. Rep. No. 103-827, at 13 (1994), reprinted in 1994 U.S.C.C.A.N. 3489 ("House Rep."). (See Brief of Petitioners USTA, CTIA, and CDT ("USTA Br.") at 7, 15.)

Two structural aspects of CALEA confirm this conclusion. CALEA's capability requirements may be enforced only in the context of a court order authorizing "an interception

[a term carefully defined at 18 U.S.C. § 2510(4)] or . . . a pen register or a trap and trace device." 18 U.S.C. § 2522(a). Section 2522 makes no reference to any other authority for obtaining information, including either of the two alternatives from which the government now seeks to fashion additional CALEA requirements. (See, e.g., DOJ Br. at 9-11 (citing a transactional records provision, 18 U.S.C. § 2703, and the general power of the courts under Federal Rule of Criminal Procedure 41).) This specification of the types of surveillance for which a court may enforce CALEA plainly implies that the capability requirements do not extend to other types of surveillance. See, e.g., Halverson v. Slater, 129 F.3d 180, 185 (D.C. Cir. 1997).

Furthermore, Congress amended only two statutes to link them to CALEA's capability requirements: Title III, 18 U.S.C. § 2518(4), and the pen register portion of ECPA, 18 U.S.C. § 3124(f), both of which now authorize a court to issue "an order . . . to enforce the assistance capability and capacity requirements under [CALEA]" in connection with a particular surveillance operation. CALEA § 201(b), Pub. L. No. 103-414, 108 Stat. 4289, 4290 (1994). CALEA did not similarly amend any other statute, including those on which the government now relies.

The government offers no support for its claim that CALEA's requirements for near-real-time isolation and delivery apply to every type of information law enforcement may have received in the past or to which it might conceivably be entitled.  CALEA requires only that carriers be able to deliver the two categories of information specifically subject to electronic surveillance laws — call content and call-identifying information. Both sources of legal authority mentioned by the government's briefs — the transactional records provision, 18 U.S.C. § 2703, and Federal Rule of Criminal Procedure 41 — concern the retrieval of stored information, not electronic surveillance or the real-time delivery of information. And the fact that pen registers or Title III intercepts in the past may incidentally have captured non-call-identifying signaling information does not transform such signals into "call-identifying information" under CALEA. 

2. Contrary to the Government's Assertions, the Order Defines CALEA's Minimum Requirements and Will Result in Intrusions on Privacy That Will Not Be Subject to Later Judicial Review.

The government attempts to evade meaningful review of the Order by contending that it establishes only a "safe harbor" without defining the statute's requirements and that any actual intrusion on privacy can be remedied on review of individual surveillance orders. (See, e.g., DOJ Br. at 3-5; FCC Br. at 31-33.) Neither argument withstands analysis.

In arguing that the Order does not itself impose any requirements, the government plainly misreads CALEA and disregards the real-world implications of the Order. The government acknowledged below that the "safe harbor" embodies the minimum set of capabilities carriers must provide and that the Order "require[s] . . . carriers to make the six [added] punch list capabilities available" to law enforcement. Order ¶ 129 (emphasis added); see also Comments of FBI at ¶ 28 (May 20, 1998) ("[T]o the extent that the Commission's standards identify statutorily required capabilities, those standards will indeed be binding on the industry." (emphasis added)). (J.A. __.) Moreover, CALEA imposes a duty on manufacturers to design equipment that implements the capabilities specified in the standard. 47 U.S.C. § 1005(b). CALEA's penalties of $10,000 per day per wiretap create a powerful incentive for carriers to avoid any departure from the Order's "safe harbor." Thus, the capabilities mandated by the Order will directly determine the information that carriers actually deliver to law enforcement.

Nor will the intrusive aspects of the FCC's Order — in particular, law enforcement's ability to capture content with a pen register — be avoided through case-by-case judicial review of surveillance requests. Pen registers are issued in ministerial, ex parte proceedings, where the court has no discretion to deny or to modify a facially valid request. 18 U.S.C. § 3123. Carriers may not object to an order on any substantive grounds. See, e.g., In the Matter of Communications Assistance for Law Enforcement Act, Report and Order, 14 FCC Rcd 4151, 4164 ¶ 29 (1999). Unlike authorized content intercepts, 18 U.S.C. § 2518(8)(d), the government has no obligation to notify an individual who was the subject of pen register surveillance, even after the close of an investigation. Thus, except in the rare case in which a prosecutor improperly tries to introduce into evidence call content received in response to a pen register order, pen register abuse will generally evade review.

CALEA's express language, legislative history, and statutory context establish that the plain meaning of "call-identifying information" is the telephone numbers used to route calls. (USTA Br. at 13-16.) The government's briefs, like the Order, are devoid of any statutory analysis of "call-identifying information" or of the key terms in its definition: "origin," "destination," "termination," and "direction." And they wholly fail to articulate a consistent framework for what "call-identifying information" means or what it encompasses. The FCC manages only a plea for deference to its naked pronouncement that "‘call-identifying information' is not limited to telephone numbers." (FCC Br. at 17, 20.) But the agency has provided no affirmative interpretation of the statutory terms to which this Court could possibly defer. (See USTA Br. at 12-13.) The FCC has offered only a smattering of multiple, shifting interpretations that assign different meanings to these terms according to the convenience of law enforcement. That approach removes any meaningful limit on the agency's discretion and does not comport with CALEA's narrow requirements.

1. The Government Cannot Overcome Petitioners' Demonstration That "Call-Identifying Information" Is Limited to the Numbers Used To Route Calls.

   Instead of offering an affirmative explanation of what "call-identifying information" means, the government unsuccessfully tries to cast doubt on petitioners' showing that call-identifying information is limited to telephone numbers dialed or transmitted to route calls.

   The FCC begins by misstating the legislative history on which petitioners rely. The agency wrongly asserts that the House Report's explanation of the meaning of "call-identifying information" refers to an earlier bill containing a different term ("call setup information") with, it says, a narrower definition than that which Congress ultimately adopted. (FCC Br. at 21.) In fact, the bill quoted and discussed in the House Report is identical to CALEA in its use of the term "call-identifying information" and, just like the enacted statute, defines that term as "dialing or signaling information that identifies the origin, direction, destination, or termination" of each communication. House Rep. at 2, 38. Indeed, the language in the House Report explaining what dialing and signaling information CALEA encompasses ("the numbers dialed or otherwise transmitted for the purpose of routing calls") appears in a paragraph that begins, "[t]he term call-identifying information means . . . ." Id. at 21 (emphasis added).

   The references in the Joint Hearings to "call setup information" do not imply that Congress later expanded the scope of CALEA. The subsequent House Report flatly states that "call-identifying information" encompasses only telephone numbers. Moreover, FBI Director Louis Freeh explained in the Joint Hearings that law enforcement was asking only for the "telephone numbers which are being called." (USTA Br. at 15 (citing Joint Hearings at 33, 50 (testimony of Director Freeh)).) Congress had no reason, and the government points to none, to give law enforcement more than it was seeking, particularly given that Congress was trying to draw a careful balance between law enforcement interests and public policies favoring privacy and minimum financial burdens on carriers and consumers.

   The changes during the legislative process to which the FCC points (FCC Br. at 21) have a simple explanation. They clarify that "call-identifying information" includes the phone numbers used to route a call through multiple switches when call forwarding is used. "Signaling information" was included in the definition of call-identifying information because, when a call is forwarded from the called party to the forwarded number, the caller does not "dial" the number to which the call is forwarded. That ultimate number is simply "signaling information" within the network and might have been excluded by a definition that referred only to "dialing information." Similarly, the terms "direction" and "termination" do not convey the additional meaning the FCC suggests, but simply capture the additional telephone numbers implicated in the context of call forwarding. (See USTA Br. at 14 (citing J-Standard).)

   Finally, contrary to the FCC's argument (FCC Br. at 21-23), call-identifying information is congruent with — and not broader than — the information law enforcement may lawfully compel with a pen register or trap and trace order. See supra at 2-4. And the plain language of the pen register statute demonstrates that this authority extends only to telephone numbers. See 18 U.S.C. §§ 3127(3)-(4); see also Brown v. Waddell, 50 F.3d 285, 293 (4th Cir. 1995) (a device "surely would have to be limited to raw telephone numbers to retain pen register status").

2. The Government Offers No Reasoned Statutory Basis for Concluding That Location Information and the Four Added Capabilities Are "Call-Identifying Information."

   The FCC fares no better in its discussion of location information and the four added capabilities. In each case, the FCC substitutes proclamations for analysis, a process that yields unexplained, shifting definitions of the key statutory terms.

   1. Location Information

   The government does not even respond to petitioners' demonstration that location is not encompassed within CALEA's definition of call-identifying information. Its principal answer is the non sequitur that law enforcement "can (and does) obtain location information" under various sources of legal authority, citing Title III, Federal Rule of Criminal Procedure 41, and the provision covering telecommunications billing records, 18 U.S.C. § 2703. (DOJ Br. at 19.) But the fact that law enforcement might sometimes receive information concerning a caller's location does not make location information a surveillance capability required by CALEA.

   The government skips over petitioners' showing that location information falls outside the plain language of CALEA because it is not associated with "each communication" and is not transmitted within the network for the purpose of routing calls. Nor does the government attempt to refute that requiring carriers to provide location information twists the meanings of "origin" and "destination" and interprets them differently and asymmetrically in similar contexts. (USTA Br. at 19-20.)

   Instead, the government contends that, if call-identifying information did not include location information, CALEA's express exclusion of location from pen register information would be "mere surplusage." (FCC Br. at 36-37.) See also Order ¶ 44 n.95 (quoting 47 U.S.C. § 1002(a)(2)(B)). But that interpretation contradicts the plain statutory language and its legislative history. (See USTA Br. at 21-22.) The explicit prohibition against delivering location information ends the previous practice of incidentally reporting such information (when available) in response to a pen register. Accordingly, the language affirms the limitation of pen registers to telephone numbers and demonstrates heightened congressional concern over the sensitivity of mobile phone location information.

   The government is equally off the mark when it suggests that mobile phone location is equivalent to the location information decipherable from a wireline telephone number, which CALEA permits carriers to provide. 47 U.S.C. § 1002(a)(2)(B) (carriers may provide location information "to the extent that the location may be determined from the telephone number" (emphasis added)). This provision was necessary to preserve the basic requirement to provide telephone numbers against an objection that such numbers impermissibly reveal location because location information is inherent in the phone numbers of some wireline telephones. That rationale is irrelevant in the context of mobile telephones. Thus, the background rule under CALEA is that location information may not be provided except when it is inherent in a telephone number.

   2. Dialed Digit Extraction

   The FCC concedes that many post-cut-through digits (e.g., bank account numbers and PINs) constitute "content" and does not contest that telephone numbers dialed by a caller after using an 800 number to reach a long distance carrier also constitute "content" from the perspective of the local carrier. (FCC Br. at 23, 32-34.) Unlike the digits that local carriers use to route calls, which appear on the "signal" channel, post-cut-through digits are found on the "content" channel.

   The FCC's argument that a carrier nevertheless must provide all information that another carrier may use to route calls proves far too much. (See id. at 23-24.) By its logic, a carrier could be required under CALEA to go into the content channel and extract not only all dialed digits, but also any spoken digits — whether to an operator who is assisting in a long distance call or to an automated voice recognition unit used by a calling card provider. Even the FBI has not claimed that CALEA goes this far.

   The only reasonable interpretation of CALEA is that "call-identifying information" encompasses information used by the assisting carrier itself. To the extent that other carriers use post-cut-through digits as "call-identifying information," law enforcement can obtain that information directly from them: CALEA was "not intended to guarantee ‘one-stop shopping' for law enforcement." House Rep. at 22. Rather, it was designed only to allow law enforcement to obtain call-identifying information "reasonably available" to the assisting carrier. Contrary to the FCC's position, that standard requires evaluating the information from the assisting carrier's perspective. For that carrier, all post-cut-through digits constitute content, not call-identifying information.

   The FCC acknowledges that its Order requires delivery of post-cut-through digits that are clearly content when law enforcement has only a pen register authorization. (FCC Br. at 32-34.) Even if some post-cut-through dialed digits are not content, where content and non-content information are inextricably intertwined, law enforcement must satisfy the higher Title III standard to obtain that information. It cannot bootstrap pen register authority into permission to receive content without complying with Title III or the Constitution, even where it has a duty to try to limit the use of the intercepted information. See infra Section IV.

   3. Hold/join/drop on Conference Calls

   The FCC offers no rationale for concluding that party hold/join/drop information is "call-identifying information." While conceding that this information goes beyond telephone numbers, the FCC argues that a conference call can have multiple origins. (FCC Br. at 24.) That is certainly true to the extent that, when different individuals call in to a conference call, each individual's phone number will be the origin of a communication. But the FCC's conclusion results not just in multiple origins, but multiple meanings of the term "origin." Under the FCC's imaginative approach, "origin" means the caller's phone number in one context, the signal indicating a party is on hold the next, and the signal indicating a party has re-joined a call in another. (And, according to the FCC, if the intercept subject happens to be on a mobile phone, origin also includes his physical location at the beginning and at the end of the call.)

   The government also does not attempt to reconcile the fact that this capability collects the identities of the parties on a call at any time with the Supreme Court's statement that pen registers do not reveal this information. United States v. New York Tel. Co., 434 U.S. 159, 167 (1977).

   4. Subject-Initiated Dialing and Signaling Information

   The FCC cannot support its assertion that subject-initiated dialing and signaling information constitutes "call-identifying information." The FCC claims that call waiting and call forwarding establish multiple "separate communications within a call." (FCC Br. at 26.) Although call waiting involves two phone calls, each with an origin and termination, it does not involve the origin and termination of a new communication every time a party switches from one call to another. Rather, the well-established understanding in the communications industry — undisputed by the FCC — is that the initial call does not terminate when a party temporarily switches to another call using call waiting. (USTA Br. at 25-26.) Similarly, the FCC offers no explanation for why call forwarding involves more than one call or communication. In any event, the FCC cannot justify shifting yet again the meanings of the terms "origin" and "termination" to force these aspects of a call into the category of "call-identifying information."

   The FCC also fails to explain how the "activation" of call forwarding service "identifies" any particular call within the meaning of CALEA, given that a subscriber may activate call forwarding without ever forwarding any call. Although Congress was concerned with call forwarding (FCC Br. at 26), the J-Standard already addresses those concerns by providing the telephone numbers identifying the direction and termination of forwarded calls so that law enforcement is not left with only the number initially dialed. J-Standard at 5. (J.A. __.)

   5. In-band/out-of-band Signaling

   The FCC continues to ignore the plain meaning of CALEA when it asserts that in-band and out-of-band signaling information is "call-identifying information" because it provides information about "whether . . . a termination occurs." (FCC Br. at 26 (emphasis added).) "Call-identifying information" encompasses only information that "identifies the . . . termination of each communication." 47 U.S.C. § 1001(2) (emphasis added). If no termination occurs — as in the case of a busy signal or the ringing of an unanswered telephone call — there is no termination to identify. Moreover, pen registers do not disclose "whether the call was even completed." New York Tel. Co., 434 U.S. at 167. Thus, this capability goes beyond what law enforcement is authorized to obtain under the pen register statute, and therefore beyond what CALEA requires.

Even if location information and the information provided by the four capabilities added by the FCC were "call-identifying information," the Order still would have to be vacated because the FCC utterly failed to engage in reasoned evaluation of whether these capabilities satisfy the other explicit statutory criteria for adding them to the industry standard.

CALEA requires only that carriers be capable of providing call-identifying information that is "reasonably available" to them. 47 U.S.C. § 1002(a)(2). While acknowledging that cost is a significant factor in determining whether a particular capability is "reasonably available," the FCC simply asserted without any analysis that the cost of each added capability was "not so exorbitant as to require automatic exclusion of the capability." Order ¶¶ 75, 82, 89, 123 (emphasis added). The government's briefs do not even attempt to suggest that this entirely meaningless formulation — exorbitant, but not "so" exorbitant — constitutes reasoned decisionmaking. Moreover, the FCC's conclusion is unsupportable on the merits. As the agency essentially concedes, it failed to consider whole categories of costs associated with implementing the added capabilities, and included costs of only a subset of the manufacturers. (FCC Br. at 31.)

The FCC also failed to evaluate whether the added capabilities satisfy the five criteria that must be met before the FCC can modify or adopt an industry standard. 47 U.S.C. § 1006(b). (See USTA Br. at 30-32.) The government dismisses these statutory conditions as mere "aspirational language" for the FCC to consider and balance. (FCC Br. at 28 & n.7.) But section 1006(b) does not simply provide that the FCC should "consider" the enumerated factors. It plainly states that the FCC may establish requirements only if they meet the five listed criteria. Section 1006(b) contains no language that makes these criteria optional.

The FCC made no effort to make the requisite findings with respect to the added capabilities. It failed, for example, to determine whether any of these capabilities are "cost-effective." The FCC argues that the cost estimates it used allowed it to assess the "‘relative' cost burden that a punch list item would impose on carriers vis a vis the J-standard capabilities that the industry had already agreed to assume." (FCC Br. at 31; see also id. at 29.) Apparently, the FCC believes that if the cost of an added capability bears some unexplained ratio to the cost of the J-Standard, then it must be, ipso facto, cost-effective. But the fact that an added capability increases the cost of the J-Standard by "only" 10% says nothing about whether the capability is cost-effective. For example, if an alternative producing the same benefit would add only 5% to the cost of the J-Standard, the capability costing 10% would not be cost effective.

What is more, notwithstanding the FCC's implication to the contrary, the costs of the added capabilities are substantial: more than $300 million under the FCC's own cost estimates, which admittedly exclude relevant implementation costs. See Order ¶¶ 75, 82, 89, 123. The fact that this amount adds approximately 30-34% to the FCC's cost estimate for the J-Standard does not establish that these capabilities are cost-effective. The FCC failed to engage in the analysis necessary to determine whether the benefits of these substantial additional expenditures, and the costs and benefits of any alternatives, render the capability cost-effective.

In addition, the FCC made no mention in its Order and is wholly silent in its brief regarding whether its requirements "minimize the cost of . . . compliance on residential ratepayers." 47 U.S.C. § 1006(b)(3). And the FCC expressly acknowledges that it failed to evaluate whether its decision would "serve the policy . . . to encourage the provision of new technologies and services to the public," id. § 1006(b)(4). (FCC Br. at 29 n.9.) Even if, as the FCC asserts, there was "little comment" on this subject (id.), that would not excuse the FCC from its duty to obey the command of section 1006.

The government acknowledges that the Order contemplates the design and implementation of systems that will deliver packet-mode call content to law enforcement agencies in response to pen register or trap-and-trace orders without satisfying the applicable Fourth Amendment and Title III requirements. The FCC clearly had the option — the course it originally proposed below — to refrain from requiring any such capability until it could adequately analyze the complexities of novel packet-mode technologies. Instead, the FCC adopted an approach that turns the Fourth Amendment, Title III, and CALEA on their heads and gives law enforcement the house so that it may obtain the kitchen sink. See Edward J. DeBartolo Corp. v. Florida Gulf Coast Bldg. & Const. Trades Council, 485 U.S. 568, 575 (1988) (an agency's statutory interpretation that "would raise serious constitutional problems" is unreasonable and accordingly "[c]ourt[s] will construe the statute to avoid such problems unless such construction is plainly contrary to the intent of Congress").

The government defends this outcome by asking this Court to endorse the absurd principle that, because law enforcement is entitled to collect call-identifying information without a warrant, it may, if necessary to obtain that information, intercept constitutionally protected call content as well. But this "lesser includes the greater" approach, which would justify all manner of unconstitutional acts, cannot be squared with the law. See, e.g., Marshall v. Barlow's, Inc., 436 U.S. 307, 309, 313-15 (1978) (holding that a court order of non-constitutional stature does not permit a government inspector to enter non-public areas of a business). It also contravenes the Supreme Court's holding that a pen register is not a Fourth Amendment "search" precisely because it does not capture the content of communications. See Smith v. Maryland, 442 U.S. 735, 741-42 (1979). Once a surveillance device becomes a vehicle for the acquisition of content, it becomes subject to the Fourth Amendment, whatever label or name it is given. A fishing license cannot become a license to drain the ocean to catch a fish.

The government supposes that law enforcement's unauthorized receipt of packet contents can be cured by trusting law enforcement to take measures to "minimize" the degree of the intrusion. (DOJ Br. at 16-17.) It relies entirely on a statute providing that law enforcement should use "technology reasonably available" to restrict pen register recording and decoding to "dialing and signaling information," 18 U.S.C. § 3121(c) — a provision clearly not designed to authorize law enforcement to receive call contents outside the strictures of Title III. This statute does nothing to prevent the unauthorized interception itself, and, even if there were some effective way to preclude improper use of intercepted information, our legal system simply does not accept the notion that privacy is adequately protected by trusting law enforcement agencies not to use private information that they are not otherwise authorized to possess.

The government also twists the prophylactic exclusionary rule, designed to curb law enforcement abuse, into a license for systematic violation of Title III. The FCC thus dismisses concerns over the contemplated invasion of privacy on the ground that law enforcement may not use its ill-gotten information in a court proceeding without Title III authorization. But law enforcement may use content information in all manner of highly intrusive and injurious ways short of its introduction as evidence in court. A violation of Title III occurs not just when information is used, but when it is acquired or even attempted to be acquired. 18 U.S.C. §§ 2510(4), 2520(1), 2511(1). An unreasonable search does not become reasonable merely because its fruits may not be used in court.

Finally, the FCC cannot defend its decision to enact a temporary order despite a concededly deficient record on packet-mode communications. (USTA Br. at 34-35.) The FCC responds by misconstruing its statutory duty and arguing that it may simply defer — for the time being — to the J-Standard because no deficiency was shown. But the FCC itself conceded that the approach it adopted raised "significant technical and privacy concerns." Order ¶ 55. And under CALEA, the FCC may neither adopt requirements that fail to meet the standards in 47 U.S.C. § 1006(b), including the protection of the privacy of communications not authorized to be intercepted, nor neglect its duty to modify an industry standard that fails to meet them. The FCC's ritualistic incantation of some of the factors contained in section 1006(b) cannot conceal that the FCC's "most suitable temporary remedy available at this time" did not begin to meet the statutory tests.

The packet-mode communication portion of the FCC's Order is not merely "not perfect," Order ¶ 56, it is fundamentally and hopelessly flawed.

CONCLUSION

For the foregoing reasons, the Court should vacate all challenged portions of the FCC's Order and remand the matter to the FCC for any necessary further proceedings consistent with its opinion.

Respectfully submitted,

______________________________ ______________________________

April 4, 2000

Free Speech | Data Privacy | Government Surveillance | Cryptography | Domain Names | International | Bandwidth | Security | Internet Standards, Technology and Policy Project | Terrorism | Authentication | Right to Know | Spam

Our Mission / Get Involved / Staff / Publications / Links / Search CDT / Jobs / Action! Previous Headlines | Legislative Tracking | CDT's Privacy Policy