Probable Cause for Reverse Keyword Warrants Under a Totality-of-the-Circumstances Test: Commentary on Chief Justice Todd’s Concurrence in Commonwealth v. Kurtz

Probable Cause for Reverse Keyword Warrants Under a Totality-of-the-Circumstances Test: Commentary on Chief Justice Todd’s Concurrence in Commonwealth v. Kurtz

I. Introduction

The concurring opinion of Chief Justice Todd in Commonwealth v. Kurtz (Pa. Dec. 16, 2025) addresses one of the most controversial investigative tools to emerge in the digital age: the “reverse keyword warrant” (or “keyword warrant”) directed at Google.

The case arises from a brutal sexual assault of a woman, K.M., in a rural part of Northumberland County. With no known suspect, the Pennsylvania State Police (PSP) sought a warrant compelling Google to disclose search queries for the victim’s name and address, along with associated IP addresses, during a narrow one-week window preceding the crime. This investigative strategy—using a broad search of a private company’s data to identify unknown suspects—triggered fundamental questions under Article I, Section 8 of the Pennsylvania Constitution and the Fourth Amendment.

The Supreme Court of Pennsylvania granted review on two central issues:

  1. Whether, under Article I, Section 8, individuals have a reasonable expectation of privacy in their private internet search queries and IP addresses such that a warrant is required; and
  2. Whether the affidavit presented to obtain the Google warrant established probable cause where there was no identified suspect and no direct evidence that Google searches were used in planning or committing the crime.

The Opinion Announcing the Judgment of the Court (OAJC) (authored by Justice Wecht) evidently reached the constitutional question. Chief Justice Todd, joined by Justices Mundy and McCaffery, concurs only in the result. She would resolve the case on narrower grounds: first, by recognizing a statutory warrant requirement for accessing certain electronic information, and second, by holding that the Google warrant was supported by probable cause under longstanding “totality-of-the-circumstances” principles.

This commentary focuses on Chief Justice Todd’s concurring opinion. It explains how she:

  • Uses statutory law to avoid deciding a novel constitutional privacy question;
  • Applies traditional probable cause doctrine to an unconventional “reverse search” warrant; and
  • Responds—implicitly and explicitly—to concerns that keyword warrants resemble unconstitutional “general warrants.”

II. Summary of the Concurring Opinion

Chief Justice Todd agrees that the defendant’s convictions and the denial of his suppression motion should be affirmed, but for reasons distinct from those of the OAJC.

A. Warrant Requirement: Resolved Statutorily, Not Constitutionally

On the first question—whether Article I, Section 8 requires a warrant to obtain private search queries and IP addresses—the concurrence declines to decide. Instead, Chief Justice Todd points to 18 Pa.C.S. § 5743(a) of the Wiretapping and Electronic Surveillance Control Act, which requires a warrant for governmental access to the contents of electronic communications in storage for 180 days or less. She notes:

  • Amici (the Office of Attorney General and Pennsylvania District Attorneys Association) highlighted this statute.
  • Section 5743 appears to “mandate a warrant” in the circumstances at issue.
  • The PSP in fact sought and obtained a warrant from a neutral issuing authority.

Because a warrant was obtained in any event, Chief Justice Todd views it as unnecessary to answer the broader, “novel constitutional question” about whether search queries and IP addresses are constitutionally protected effects or “possessions” under Article I, Section 8. This is an application of constitutional avoidance: decide cases on non-constitutional grounds when possible.

B. Probable Cause: Affirming the Warrant Under Totality-of-the-Circumstances

Turning to the second issue, Chief Justice Todd engages fully: Did the affidavit establish probable cause to issue a reverse keyword warrant to Google?

She holds that it did. Relying on:

  • The secluded location of K.M.’s home;
  • The timing of the attack (when her husband was on an overnight shift);
  • The fact that the victim was transported and then abandoned roughly 0.7 miles away in another isolated location; and
  • Police experience that serious sexual offenders are often long-term, “fantasy-driven” stalkers,

she concludes that it was reasonable—and indeed “highly probable”—to infer that the assailant researched K.M. and her residence online, likely using Google, given the ubiquity of that search engine.

The warrant was also tightly cabined:

  • A brief period: one week before the rape (July 13–20, 2016);
  • Specific targets: Google search and image queries containing the victim’s name or address; and
  • Limited data: the queries and associated IP addresses (not all user content).

Viewed through the deferential, common-sense lens that governs probable cause review, she finds there was a “fair probability” that evidence of the crime would be found in Google’s records. Consequently, she rejects suppression and concurs in affirmance of the Superior Court, without joining the OAJC’s broader constitutional reasoning.

III. Background and Procedural Posture

Although the concurring opinion does not re-tell the full factual narrative (deferring to the OAJC), it provides enough context to understand the investigative choice at issue:

  • K.M. lived in a “very small rural development” near Milton, Pennsylvania—an address “likely unknown by most.”
  • She was attacked in her home at night while her husband was on an overnight work shift.
  • The assailant transported her to another location by vehicle, raped her, and then left her in a secluded area about 0.7 miles from her home.

With no identified suspect and no direct digital trail, the PSP sought a search warrant directed to Google (the “Google warrant”). Trooper Follmer’s affidavit stated, in substance, that:

  • The victim and her residence were not randomly targeted.
  • The assailant appeared familiar with the victim, her home, and her husband’s schedule.
  • Based on police experience, sexual offenders of this kind are “predominately fantasy driven” and may stalk and research victims over time.
  • “It is believed that the Actor also researched the VICTIM’S address and/or VICTIM utilizing the internet or search engine similar to the searches provided by Google Inc.”

The warrant sought:

  • Search queries and image-search queries using the victim’s name or related to her address; and
  • The IP addresses associated with those queries;
  • Over the seven days prior to the assault.

The magistrate issued the warrant. The defense moved to suppress the fruits of the Google warrant as unsupported by probable cause and argued that it functioned as an impermissible dragnet or “general warrant.” The trial court denied suppression. The Superior Court affirmed, finding that the affidavit showed a high probability that the attack was planned and that Google would likely contain relevant evidence.

On allocatur, the Supreme Court of Pennsylvania considered both the constitutional/warrant requirement question and the sufficiency of probable cause. The OAJC reached and resolved the constitutional issue. Chief Justice Todd, however, would uphold the warrant on statutory and probable cause grounds alone.

IV. Detailed Analysis

A. Precedents and Authorities Cited

The concurrence is steeped in classic search-and-seizure doctrine. Chief Justice Todd grounds her analysis in both federal and Pennsylvania authorities, using them to show that:

  1. Probable cause is a flexible, practical standard based on probabilities, not certainties; and
  2. Search warrants may serve as investigative tools—even when no suspect is known—provided probable cause exists that evidence will be found in the place to be searched.

1. Constitutional Texts and Rules of Criminal Procedure

  • U.S. Const. amend. IV & Pa. Const. art. I, § 8. Both require warrants to be supported by probable cause and to particularly describe the place to be searched and things to be seized. Article I, Section 8 uses slightly different language (“possessions,” “as nearly as may be”), but the basic structure is parallel.
  • Pa.R.Crim.P. 201 & 203. Rule 201 identifies what may be searched for and seized (contraband, fruits, instrumentalities, evidence, or persons). Rule 203(B) requires that no warrant issue without probable cause, supported by one or more affidavits sworn to before the issuing authority, and codifies the “four corners” rule:
    • The magistrate may not consider evidence outside the affidavit.
    • Later reviewing courts are likewise confined to the affidavit’s four corners.

2. The Probable Cause Standard and Totality-of-the-Circumstances

  • Illinois v. Gates, 462 U.S. 213 (1983). Gates famously abandoned rigid tests in favor of a “totality-of-the-circumstances” analysis. Probable cause is a “fluid concept” based on practical, common-sense judgments about whether there is a “fair probability” that evidence will be found in a particular place. Affidavits should be interpreted in a non-hypertechnical manner, and reviewing courts should afford “great deference” to the issuing magistrate.
  • Commonwealth v. Gray, 503 A.2d 921 (Pa. 1985). Pennsylvania formally adopted Gates’ totality-of-the-circumstances approach.
  • Commonwealth v. Edmunds, 586 A.2d 887 (Pa. 1991); Commonwealth v. Miller, 518 A.2d 1187 (Pa. 1986). Edmunds reiterates that probable cause is the “linchpin” of warrant issuance; Miller is cited for the same proposition.
  • Commonwealth v. Torres, 764 A.2d 532 (Pa. 2001); Commonwealth v. Pacheco, 263 A.3d 626 (Pa. 2021). These cases confirm that reviewing courts:
    • Do not engage in de novo re-weighing of the affidavit; and
    • Ask only whether the magistrate had a “substantial basis” for concluding probable cause existed.
  • United States v. Davis, 458 F.2d 819 (D.C. Cir. 1972). Quoted for the observation that probable cause is a “plastic concept” and a “kaleidoscopic myriad” of factors; rarely does one decision control the next. This reinforces that flexible, case-specific assessment is required.
  • Commonwealth v. Davis, 351 A.2d 642 (Pa. 1976); Commonwealth v. Rodriguez, 585 A.2d 988 (Pa. 1991). These decisions emphasize:
    • The Commonwealth must furnish sufficient information to persuade a reasonable person that probable cause exists to conduct a search.
    • Probable cause exists when facts known to the affiant would warrant a person of reasonable caution in believing a search should be conducted.
  • Commonwealth v. Baker, 518 A.2d 802 (Pa. 1986); Commonwealth v. Baker, 615 A.2d 23 (Pa. 1992). Cited for the principle that probable cause is based on probability, not on proof sufficient for conviction. A prima facie showing of criminal activity is not required at the warrant stage.
  • Commonwealth v. Edwards, 426 A.2d 550 (Pa. 1981). Directs that affidavits be read in a “common sense, nontechnical, ungrudging and positive manner.”
  • Commonwealth v. Green, 265 A.3d 541 (Pa. 2021). Reaffirms that probable cause involves probabilities, not certainties, and does not demand a prima facie case.
  • Commonwealth v. Jacoby, 170 A.3d 1065 (Pa. 2017). Appellant cited Jacoby for the proposition that probable cause requires a showing that a particular person committed a particular crime and that evidence will be found in a particular place. Chief Justice Todd responds that, while warrants typically connect specific persons or items to crimes, they can also be investigative when properly supported.

3. Warrants as Investigative Tools; “Mere Evidence” and Unknown Suspects

  • Warden, Maryland Penitentiary v. Hayden, 387 U.S. 294 (1967). Hayden held that warrants may authorise seizure not only of fruits and instrumentalities, but also of “mere evidence” of crime. Importantly, the Court underscored that it is reasonable to conduct searches to obtain evidence that will aid in apprehending and convicting criminals.
  • Commonwealth v. Butler, 291 A.2d 89 (Pa. 1972). Acknowledges that warrants must not become general tools of investigation; they require a connection between the items sought and the crime.
  • Commonwealth v. Jones, 988 A.2d 649 (Pa. 2010). Jones makes explicit that a warrant can serve as an investigative tool, even when the identity of the murderer or victim is not known. This is crucial for Chief Justice Todd’s acceptance of reverse keyword warrants aimed at identifying an unknown assailant.

4. Why We Require Warrants: Neutral Magistrate and Privacy Protection

  • McDonald v. United States, 335 U.S. 451 (1948). Cited for the fundamental rationale of the warrant requirement: placing a neutral, detached magistrate between police and citizen, so that an objective judgment can be made about the need to invade privacy.
  • Riley v. California, 573 U.S. 373 (2014). Although Riley concerned cell phone searches, it is invoked here to reinforce that warrants ensure inferences supporting a search are drawn by a neutral magistrate, not by an officer engaged in “the often competitive enterprise of ferreting out crime.”

5. Statutory Authority: 18 Pa.C.S. § 5743

Amici for the Commonwealth highlighted 18 Pa.C.S. § 5743(a), which provides that law enforcement may require disclosure of the contents of a communication in electronic storage for 180 days or less “only pursuant to a warrant issued under the Pennsylvania Rules of Criminal Procedure.” Chief Justice Todd notes:

  • This statute appears to create a statutory warrant requirement for accessing certain electronic information.
  • The PSP undisputedly obtained such a warrant here.
  • It is therefore unnecessary—at least for purposes of her analysis—to resolve definitively whether § 5743 applies to Google search queries or to decide the broader constitutional question about expectation of privacy in searches and IP addresses.

B. Legal Reasoning in the Concurrence

1. Constitutional Avoidance and the Statutory Warrant Requirement

The first move in Chief Justice Todd’s reasoning is jurisdictional and methodological: she narrows the case.

Rather than addressing whether Article I, Section 8 independently demands a warrant for Google search histories and IP addresses, she observes that the General Assembly has already legislated a warrant requirement (at least for many forms of electronic content) in § 5743. Because the PSP complied and obtained a warrant, she sees no pressing need to adjudicate the constitutional question.

This approach has several implications:

  • It embodies constitutional avoidance—courts should not reach constitutional questions if a case can be resolved on non-constitutional grounds.
  • It recognizes the General Assembly’s role in shaping privacy protections for emerging technologies.
  • It leaves open—and does not decide—the precise constitutional status of private internet search queries and IP addresses under Article I, Section 8.

By framing the case as a pure probable cause challenge to an already lawfully obtained warrant, the concurrence channels the analysis into familiar doctrinal territory rather than pioneering a new constitutional rule.

2. Re-stating the Probable Cause Standard

Chief Justice Todd spends substantial time rearticulating the governing standard:

  • Probable cause is a “fluid concept” (Gates), dependent on “the assessment of probabilities in particular factual contexts.”
  • It is evaluated by a totality-of-the-circumstances test, not rigid rules.
  • The issuing authority’s task is to determine whether there is a “fair probability” that contraband or evidence of a crime will be found in the place to be searched (Rodriguez).
  • Reviewing courts must:
    • Limit themselves to the four corners of the affidavit (Pa.R.Crim.P. 203(B));
    • Read the affidavit in a common-sense, nontechnical, and “ungrudging” manner (Edwards); and
    • Defer to the magistrate’s determination if there was a “substantial basis” for finding probable cause (Gates, Pacheco).
  • Probable cause does not require:
    • Direct evidence of criminality;
    • Elimination of all innocent explanations; or
    • A prima facie (trial-level) showing of guilt (Baker, Green).

Embedded in this restatement is an important practical point for digital investigations: the absence of direct evidence that a specific digital tool was used does not preclude probable cause, so long as it is reasonable, given the circumstances and common experience, to infer that such a tool was likely involved.

3. Application to the Google Keyword Warrant

Chief Justice Todd then applies these principles to Trooper Follmer’s affidavit. Key factual and inferential steps in her reasoning include:

  • Non-random targeting of victim and residence.
    • K.M.’s home is in a “very small rural development” whose whereabouts are “likely unknown by most.”
    • The attack occurred when her husband was on an overnight shift, and she was alone in bed.
    • The victim was dropped off close (0.7 miles) to her residence in another secluded area.
    These facts, taken together, suggested that the assailant:
    • Knew who K.M. was;
    • Knew where she lived; and
    • Knew her husband’s schedule.
  • Police experience with sexual offenders.
    • The affidavit recounted that “most sexual offenders of this magnitude are predominately fantasy driven.”
    • They may stalk a victim over a period of time, fantasizing and planning before acting.
    • Given that K.M. had moved into the area in July 2015, it was plausible that the attacker had seen or become aware of her and then followed or researched her.
  • Inference of online research using Google.
    • Given the ubiquity of internet search technology and Google’s dominance (widely recognized even colloquially—“to Google”), it was reasonable to infer that the assailant used an online search engine to:
      • Locate the victim’s residence;
      • Confirm her address; or
      • Research her and her husband’s routines.
    • The concurrence explicitly describes the inference that the attacker used Google as “reasonable” and, in Chief Justice Todd’s view, “highly probable.”
  • Targeted scope of the warrant.
    • Temporal limit: a seven-day window preceding the rape (July 13–20, 2016).
    • Substantive limit: only search queries and image-search queries:
      • Using K.M.’s name or image; or
      • Relating to her specific address.
    • Data requested: the queries themselves and the associated IP addresses.

From these components, the concurrence concludes that the affidavit established a “fair probability” that Google’s search records would contain evidence related to the crime, particularly the identity or digital traces of the assailant.

Chief Justice Todd underscores that there was:

  • No requirement that Trooper Follmer have direct evidence that Google, as opposed to some other means, was used; and
  • No requirement that he rule out other innocent uses of Google that might generate similar queries.

In short, under the totality-of-the-circumstances, and read with proper deference and common sense, the warrant was supported by sufficient probable cause.

4. Reverse Keyword Warrants and the “General Warrant” Objection

Amici for the defendant (Electronic Frontier Foundation, NACDL, PACDL, ACLU, and others) mounted a powerful critique:

  • They likened keyword warrants to the “general warrants” used by British authorities before the American Revolution, which allowed broad searches of many people’s property on the bare hope of turning up some evidence.
  • They argued that a keyword warrant is comparable to authorizing a search of “every house in an area of a town” to see whether any contain writings linked to a crime.
  • They stressed that such warrants are “reverse searches”:
    • Instead of targeting a known suspect, the government asks a third-party (Google) to sift through its entire user database to find all accounts that meet certain query criteria.
    • This “traps countless innocent people in its net.”
  • They maintained that:
    • The warrant lacked particularized probable cause as to each user whose data was searched; and
    • Probable cause must be shown “as to every user” swept into such a reverse search.

Chief Justice Todd does not directly re-theorize what “general warrants” mean in the digital context. Instead, she responds within the existing probable cause and particularity framework:

  • She emphasizes that:
    • Probable cause concerns the likelihood that evidence will be found in the place to be searched—here, Google’s records matching tightly defined criteria.
    • Search warrants may be used as investigative tools when properly tailored (Hayden, Jones), especially where a suspect’s identity is unknown.
  • She underscores the warrant’s narrow scope—limited timeframe and specific keywords (the victim’s name and address)—as opposed to a broad trawl through random Google data.
  • She reiterates that the law does not require:
    • Probable cause “as to every user” whose queries might be incidentally reviewed; or
    • Elimination of all innocent explanations or non-criminal possibilities.

By affirming the Google warrant as sufficiently supported by probable cause and distinguishing it from impermissible general warrants, the concurrence effectively accepts the concept of reverse keyword warrants under Pennsylvania law, at least when cabined by:

  • Substantial factual support from the circumstances of the crime;
  • Experienced-based inferences about typical offender behavior; and
  • Tight limitations in time and scope.

5. Unknown Suspects and “Dragnet” Concerns

Appellant argued that:

  • The PSP had “no suspect, no leads, and no evidence” and thus took a “shot in the dark.”
  • The warrant functioned as a “dragnet” for any and all people who searched for K.M. or her address, many or most of whom would be innocent.
  • Without identifying a suspect or direct evidence that Google was used, the warrant lacked a factual anchor.

Chief Justice Todd’s answer has three components:

  1. Investigative warrants are permissible. Citing Jones and Hayden, she reiterates that warrants can serve to gather evidence when a suspect’s identity is unknown. A warrant need not always be directed at a named individual; it can target a location or repository of potential evidence (here, Google’s servers) if supported by a sufficient factual nexus.
  2. Probable cause deals in probabilities, not certainties. At the warrant stage, the police are not required to prove that Google was used—only that it was likely used, given the circumstances, and that evidence probably lies in Google’s records. Inferences from police experience and modern realities of technology use are legitimate ingredients in that probability assessment.
  3. Targeted scope distinguishes this from a pure dragnet. The warrant did not authorize search of all users’ data or all search terms, but a defined subset:
    • A short temporal span;
    • Very specific search terms (victim’s name and address); and
    • Limited types of information (search queries and associated IP addresses).

Accordingly, the concurrence rejects the “fruit of the poisonous tree” argument: because the warrant was valid, all evidence derived therefrom is admissible.

V. Impact and Significance

A. Implications for Digital Search Warrants in Pennsylvania

Although a concurrence does not speak for the full Court, Chief Justice Todd’s reasoning—joined by two justices—will be influential in lower courts confronted with similar digital warrants, particularly:

  • Keyword warrants. The opinion signals that reverse keyword warrants are not per se unconstitutional or invalid for lack of probable cause. Instead:
    • Their validity turns on the totality of the circumstances—the relationship between the crime, the facts known to police, and the particular digital data sought.
    • Well-cabined keyword warrants, backed by plausible inferences about likely offender behavior and technology use, can satisfy probable cause.
  • Geofence and other reverse-location warrants. While not directly addressed, the same reasoning could be analogized to warrants seeking anonymized location data for devices within a particular area during a crime window. The concurrence’s emphasis on:
    • Investigative warrants for unknown suspects;
    • Probability-based inferences; and
    • Temporal and spatial narrowness;
    provides an analytical template likely to be invoked by the Commonwealth in such cases.
  • Role of police experience and technological ubiquity. The concurrence normalizes reasoning such as: “Because Google (or similar tools) are ubiquitous and widely used, there is a fair probability a reasonably sophisticated offender would have used them.” That form of inference—once controversial—gains some doctrinal support here.

B. Balancing Privacy and Investigative Needs

The amici’s concerns are significant: keyword warrants inherently sweep in data of many innocent users. The concurrence’s answer is essentially doctrinally conservative: the same probable cause and particularity standards govern, but they are applied flexibly to new contexts.

Several privacy-related consequences follow:

  • No categorical bar. Individuals cannot rely on Article I, Section 8—at least as interpreted in this concurrence—to categorically invalidate reverse keyword warrants. Arguments must instead focus on:
    • Whether the affidavit truly shows a sufficient factual nexus between the crime and the digital data sought; and
    • Whether the warrant’s temporal and subject-matter scope is sufficiently narrow.
  • Importance of tailoring. The tighter the warrant’s limits (short time window, specific search terms directly tied to a specific victim or event, restricted categories of data), the more likely it will survive a probable cause and particularity challenge. Overbroad keyword warrants—long time periods, generic keywords, or expansive data categories—may still be vulnerable.
  • Ongoing tension with the “general warrant” analogy. The concurrence does little to directly settle whether, and when, a reverse keyword warrant becomes functionally indistinguishable from a general warrant. Future cases will have to wrestle with the outer limits: How many users can a warrant implicate before it becomes constitutionally suspect? How generic can search terms be?

C. Interaction with the OAJC and Article I, Section 8 Jurisprudence

The OAJC, by addressing the first issue, appears to set forth a substantive Article I, Section 8 rule regarding privacy in search queries and IP addresses. Chief Justice Todd’s concurrence stands as a counterpoint in terms of judicial method: she would rely on statutory law and traditional probable cause doctrine rather than creating or extending constitutional protections in this specific case.

In the longer term, this highlights two competing approaches in Pennsylvania search-and-seizure jurisprudence:

  • Constitution-forward. Developing distinct, sometimes more protective rules under Article I, Section 8 (as in Edmunds) for emerging technologies and data types.
  • Statute-and-probable-cause forward. Emphasizing existing statutory protections (like § 5743) and applying flexible, but traditional, probable cause principles to new investigative methods without dramatically altering constitutional doctrine.

The concurrence clearly aligns with the latter approach in this case, reinforcing legislative primacy in defining many aspects of digital privacy and leaving the constitutional baseline less fully elaborated.

D. Practical Guidance for Courts and Practitioners

For trial judges and practitioners evaluating or litigating reverse digital warrants, the concurrence suggests several practical guideposts:

  • Affidavits must do more than recite technology’s ubiquity. They should:
    • Describe concrete facts of the crime suggesting planning, stalking, or research;
    • Connect those facts to likely online behaviors (e.g., searching for a victim’s name or address); and
    • Explain, based on training and experience, why an offender is likely to use a particular kind of digital tool.
  • Scope matters greatly. Narrow timeframes and specific, crime-linked keywords are much easier to uphold than broad, exploratory keyword sets or long windows.
  • Framing the “place to be searched.” Especially with reverse warrants, it is critical to frame the place as the company’s records falling within narrowly specified parameters, not the entirety of an online platform.
  • Argument focus. Defense challenges will be most effective if they:
    • Show that the affidavit’s inferences are too speculative in the case-specific context; or
    • Demonstrate that the warrant’s parameters approximate a general search by being insufficiently particular.

VI. Complex Concepts Simplified

A. Probable Cause

“Probable cause” means there is a reasonable likelihood—based on facts and reasonable inferences—that:

  • A crime was committed; and
  • Evidence of that crime will be found in the place to be searched.

It does not mean proof beyond a reasonable doubt or even enough evidence to convict. It is a commonsense judgment about probabilities, not certainties.

B. Totality-of-the-Circumstances Test

Under this test, courts do not apply strict formulas. Instead, they:

  • Look at all the facts in the affidavit together;
  • Consider how those facts fit with one another;
  • Ask whether a reasonable person would think there is a fair chance evidence will be found; and
  • Give weight to the officer’s experience and to realistic views of how people behave.

C. Four Corners Rule

When determining if a warrant was supported by probable cause, courts may only look at the information contained within the four corners of the written affidavit presented to the magistrate. They cannot consider:

  • Unsworn oral explanations; or
  • Information known to police but omitted from the affidavit.

D. Reverse Keyword (or Keyword) Warrant

A “keyword warrant” is a type of “reverse search” where:

  • Police do not start with a known suspect.
  • Instead, they ask a company like Google to search its database for:
    • Any user who searched for specific terms (e.g., a victim’s name or address);
    • During a particular time period; or
    • From certain locations.
  • The goal is to use these search records to identify suspects.

This contrasts with traditional warrants, which focus on a particular suspect, device, or place linked by probable cause to a crime.

E. General Warrants

“General warrants” are broad search authorizations historically used by British authorities to rummage through people’s homes, papers, and effects without specific limits. The American Revolution and subsequent constitutions (federal and state) were, in significant part, a reaction against such general warrants.

Modern law prohibits warrants that:

  • Are not based on probable cause; or
  • Fail to particularly describe what is to be searched and what is to be seized.

The controversy surrounding keyword warrants is whether, even with some constraints, they are so sweeping in their capture of innocent data that they function like modern general warrants.

F. Fruit of the Poisonous Tree

This doctrine holds that:

  • If evidence is obtained through an unconstitutional search (the “poisonous tree”), then evidence later derived from that illegality (the “fruit”) is generally inadmissible.

In Kurtz, the defendant argued that if the Google warrant was invalid, then all evidence that flowed from it should be suppressed. Chief Justice Todd rejects this argument by concluding the warrant was supported by probable cause and therefore not “poisonous.”

VII. Conclusion

Chief Justice Todd’s concurring opinion in Commonwealth v. Kurtz does not set out the full constitutional contours of privacy in internet search queries and IP addresses under Article I, Section 8. Instead, it does something equally important for the day-to-day administration of justice: it shows how traditional probable cause doctrine can be applied to novel digital investigative techniques like reverse keyword warrants.

Key takeaways from the concurrence include:

  • Statutory protections matter. The presence of a statutory warrant requirement for electronic communications (18 Pa.C.S. § 5743) can allow courts to avoid deciding expansive constitutional questions.
  • Reverse keyword warrants are not categorically invalid. When:
    • The affidavit links the crime to likely online research;
    • Relies on realistic assessments of offender behavior and technological ubiquity; and
    • The warrant is tightly tailored in time and subject matter;
    probable cause can exist to search a platform’s records even when the suspect is unknown and there is no direct proof of digital use.
  • Probable cause remains a pragmatic, probability-based standard. Courts review affidavits with common sense and deference to issuing magistrates, not through hypertechnical reconstruction or demands for direct evidence of every inference.
  • Privacy and general-warrant concerns persist. While the concurrence accepts reverse keyword warrants in this context, it does not eliminate the underlying tension between broad digital searches and the historical prohibition on general warrants. The boundary lines—especially concerning scope and particularity—will continue to be contested in future litigation.

In the broader legal context, the concurrence exemplifies a cautious, doctrine-grounded response to the challenges of digital evidence. It affirms that existing search-and-seizure principles are flexible enough to address new technologies, while also signaling that legislative choices and careful warrant drafting will be central to maintaining a constitutionally acceptable balance between investigative effectiveness and individual privacy.

Case Details

Year: 2025
Court: Supreme Court of Pennsylvania

Judge(s)

Todd, Chief Justice Debra

Comments