Incriminating thoughts.

• Author: Farahany, Nita A.

INTRODUCTION I. THE DEVELOPMENT OF THE TESTIMONIAL/PHYSICAL DICHOTOMY A. Early Conflict over the Scope of Privilege B. Schmerber v. California: A Codified Divide C. Rationalizing Privilege 1. The excuse-based model of privilege II. THE SPECTRUM FROM NEUROSCIENCE A. Identifying B. Automatic 1. Autonomic (functional impairment) 2. Visceral (emotional state) 3. Visceral (behavioral traits) C. Memorialized. 1. Recognition a. Voices and faces b. Guilty knowledge 2. Episodic memories 3. Papers and memorialized evidence a. Categorizing papers b. Reconciling papers and other memories D. Utterances 1. Voluntary utterances 2. Evoked utterances a. Active memory recall b. Binary responses c. Complex words and thoughts III. RECOGNIZING COGNITIVE LIBERTY A. Riddles Revealed B. Discomforting Thoughts CONCLUSION INTRODUCTION

The neuroscience revolution poses profound challenges to current self-incrimination doctrine and exposes a deep conceptual confusion at its core. In Schmerber v. California, the Court codified the basic dichotomous principle that the Self-Incrimination Clause ensures that no person shall be forced to give self-incriminating testimony in a criminal case, but may be compelled to provide real or physical evidence. (1) The Fifth Amendment protects against "coercion [to] prove [a] charge against an accused out of his own mouth," (2) but not extracting saliva from the same. (3)

The dichotomy between testimonial and physical evidence has failed to achieve its simplifying purpose of making easier determinations of when Fifth Amendment privilege applies. And it opens the door to a future in which the government might easily target and obtain the hidden contents of the mind--from undisclosed thoughts to ideas and emotions. Scholars and practitioners lament the impracticability of a testimonial/physical dichotomy, (4) lambast its flawed historical roots, (5) and introduce a myriad of new approaches for respecting its purpose. (6) This Article seeks to reframe the debate over what "to be a witness" means in the context of self-incrimination. It demonstrates both historically and through modern examples from neuroscience the inadequacy of defining "to be a witness" through a testimonial/physical dichotomy. It proposes a new taxonomy of evidence that squares prior case law with the predominant descriptive rationale underlying the privilege. Yet in so doing it foretells a discomfiting fate for a sphere of mental privacy.

The following fictitious scenario provides a framework to understand how neuroscience reveals the need for a new taxonomy underlying the privilege against self-incrimination:

A woman is murdered in her home, and the crime captured on a homesecurity video recording. The recording reveals the woman struggling with an unidentifiable masked man holding a hammer with a uniquely patterned handle. In the ensuing struggle, the woman delivers a dizzying blow to the back of his head. A second masked man enters the room, and says to the first, "Let's go!" The second man uses the hammer to fatally wound the woman. The video ends with the two men dragging the woman out of the home. When the police arrive at the crime scene the next day, they find no trace of the crime save the videotape and the hammer. The police correctly suspect that the woman's husband is one of the perpetrators and two weeks later bring him in for questioning. Now in the police interrogation room and faced with the husband, an uncooperative suspect who sits silently before them, the police contemplate methods by which they could obtain further evidence from him. At their disposal are the following techniques: (1) obtain structural images of the husband's brain to see if he suffers from head trauma in the region where the perpetrator suffered a blow to the head; (2) measure his automatic emotional and physiological responses to stimuli such as a photograph of his wife; (3) obtain his spoken, written, or stored memories of the night of the crime; or (4) elicit brain-based but interpretable responses to their questions by whatever means necessary, including torturous ones.

Each technique the police could use would yield tangible physical evidence detectable through modern neuroscientific methods. The police could use such evidence to make inferences about the husband's involvement in the crime. Faithful application of the testimonial/physical dichotomy should therefore predict that the police could compel any of the brain-based evidence without running afoul of the Fifth Amendment. And yet something seems amiss with this prediction--it fails to capture intuitions about mental privacy and autonomy of self.

Other scholars have already recognized the potential conflict between emerging neuroscience and current self-incrimination doctrine. (7) That scholarship has been limited by its narrow focus on particular neuroimaging technologies and by an understandable attempt to fit new neuroscience into old categories. (8) Even the more nuanced discussions have included only stimuli-based imaging without considering the myriad of other applicable neuroscientific technologies at issue. This narrow approach makes suspect the predictions these scholars make for how certain neuroimaging technologies would fare under current doctrine. (9) Using outmoded and historically rejected rationales such as mental privacy (10) these scholars divide on whether neuroimaging is physical and unprivileged or testimonial and privileged. (11)

This Article instead takes a more integrated view of cognitive neuroscience to bring coherence and restore meaning to self-incrimination doctrine. An integrated approach reveals that evidence can arise from identifying characteristics concerning individuals; it can arise automatically, without conscious processing; it can arise through memorialized photographs, papers, and memories; or it can arise as uttered responses, whether made silently or aloud. A spectrum of evidence that spans identifying, automatic, memorialized, and uttered evidence offers more nuance and precision than the traditional testimonial/physical dichotomy, and gives descriptive power to the rationale underpinning the privilege. That rationale, the excuse model of the privilege, provides a constitutional excuse for a suspect's silence when failing to do so would put the defendant in the difficult position of choosing between lying, facing contempt, or incriminating himself.

But the spectrum also does more. The excuse-based model, now imbued with greater descriptive force, reveals that current doctrine is misaligned with moral intuitions about mental privacy. The limited excuse for silence has been described as balancing societal interests in securing truth in criminal prosecutions, in promoting law-abiding behavior, and in preserving the legitimacy of the criminal justice system. Although descriptively robust, that limited excuse may simply be too weak to protect individuals against self-incrimination in the post-neuroscientific era.

This Article begins with the doctrinal underpinnings of the testimonial/physical divide, including Schmerber v. California and its progeny. Because this landscape is already well groomed, (12) the discussion here focuses on the nuances of the dichotomy and how it aligns with an excuse-based rationale of privilege. Part II then uses modern examples from neuroscience to reveal the impracticability of the testimonial/physical dichotomy and its inapplicability to emerging technology. Neuroscience offers more than a mere axe to cleave the divide. It provides an analogical tool by which a new taxonomy of evidence can be constructed. Along that spectrum of evidence, past cases and future ones are located, which both explains current doctrine and predicts its evolution. The spectrum runs from identifying, to automatic, memorialized, and finally uttered evidence. Realigning self-incrimination privilege along this new spectrum instead of the testimonial/physical dichotomy brings coherence to existing doctrine and enables a more meaningful prediction of how emerging neuroscience will fare. That prediction--that identifying, automatic, and even voluntarily memorialized evidence go beyond the scope of privilege--motivates a timely discussion in Part III about how an excuse-based model of privilege fails to reflect modern intuitions about the purpose of the Self-Incrimination Clause. Part III concludes by finding that although other constitutional guarantees may provide some protection for cognitive liberty, it is unlikely that our existing constitutional structure will ensure a sphere of mental privacy that will safeguard individual autonomy.

1. THE DEVELOPMENT OF THE TESTIMONIAL/PHYSICAL DICHOTOMY

   1. Early Conflict over the Scope of Privilege

      Well before Schmerber v. California (13) or the incorporation of the Self-Incrimination Clause of the Fifth Amendment against the states, courts disagreed about when a criminally accused person could be compelled to be a witness against himself. Some state courts believed the privilege guarded against the compulsion of any form of self-incriminating evidence, while others found it applicable only to testimonial statements.

      States that interpreted the privilege broadly held that an accused could not be forced to exhibit his physical features to a jury, (14) to try on articles of clothing before the jury, (15) to submit to a physical medical examination, (16) or to match a footprint impression found at the crime scene. (17) By contrast, state courts that interpreted the privilege more narrowly found it permissible to compel an accused to grow a beard, (18) to try on articles of clothing before the jury, (19) to provide handwriting exemplars, (20) to submit to physical examination, (21) to provide fingerprints (22) or other exemplars, or to illustrate the fit of his foot to a crime scene footprint. (23)

      The disagreement was fundamental to the purpose and the scope of the privilege against...