A. Introduction
Technology is now present in almost every aspect of life. The criminal justice system is no exception.
However, the use of new technologies in this specific context is the source of many debates about the possible violation of fundamental rights of individuals. For example, the legitimacy of robot judges and robot lawyers,Footnote 1 the legality of using predictive models to anticipate the commission of crimes,Footnote 2 the limits on obtaining personal information from the social media of the defendant,Footnote 3 and the possibility of using technological systems to assess the dangerousness of individuals,Footnote 4 determine their guiltFootnote 5 or the sanction to be applied,Footnote 6 among others, are being debated.
One type of new technology that has been discussed recently is brain-reading devices. Consider the following case: A person is the prime suspect in a murder investigation. In addition to forcing that person to submit to a blood test and hand over some documents, a judge also issues a warrant forcing the accused to submit to an examination by a brain-reading machine so that the prosecution can collect biometric and neurological information, which is then translated into a reading of mental states and thoughts.Footnote 7
The situation described above is no longer a science fiction hypothesis, as brain-reading technologies already exist and are even used in criminal investigations in some countries, such as Japan.Footnote 8 According to Matsuda, Ogawa, and Tsuneoka, there are about 100 people in Japan who have been trained as examiners at the Forensic Science Training Center, an institution affiliated to the National Institute of Police Science Research. These examiners handle about 5,000 cases a year.Footnote 9
As humanity has been trying to develop a mechanism to detect lies since ancient times,Footnote 10 the possibility that brain-reading technologies can distinguish truth from lies has created great expectations. However, their existence has also raised fears that the authorities could use them to read people’s minds without their consent and obtain information that could criminally incriminate them. This possibility immediately poses questions about the right against self-incrimination. Could a person charged in a criminal case invoke his right against self-incrimination and legitimately refuse to submit to a brain-reading test? Could a court order him to take the test or punish him for refusing?
The purpose of this Article is to analyze whether the collection of incriminating information through the non-consensual use of brain-reading technologies violates the right against self-incrimination under its traditional interpretation, according to which the scope of application of this right includes only “testimonial evidence,” thus excluding “real or physical evidence.” It must be noted therefore that the Article does not address the possibility that the non-consensual use of brain-reading technologies can violate other rights, such as the rights to privacy and to freedom of expression.
In light of the above, this Article is not intended to critically analyze or support this interpretation,Footnote 11 nor is it intended to propose an alternative one, at least not in depth. On the contrary, the purpose of this Article is more modest: To illustrate the difficulties and challenges that brain-reading technologies raise for the aforementioned traditional interpretation of the right against self-incrimination. Because of this, the Article can be considered inquiring and descriptive. However, even if the stated purpose could be considered modest, I believe it can be useful because it is quite likely that new and more effective brain-reading technologies will be developed in the future. Therefore, the relevant questions should be discussed and hopefully answered as soon as possible.
The Article is structured as follows: The first part describes and explains the performance of three currently available brain-reading technologies: Functional magnetic resonance imaging, P300, and semantic decoding. The second part addresses the problem of the scope of application of the right against self-incrimination, in particular what kind of evidence is covered by this right. As stated, the traditional interpretation of this right argues that its scope includes only “testimonial evidence,” thus excluding “real or physical evidence.” This interpretation has been adopted by the United States Supreme Court, the Canadian Supreme Court, the Constitutional Court of Spain, the Constitutional Court of Chile, and the ECHR. These courts have been chosen to illustrate that, even in different legal systems, a remarkably similar interpretation of the right against self-incrimination has developed. Finally, in the third part an attempt will be made to apply this traditional interpretation of the right against self-incrimination to cases of non-consensual collection of incriminating information through the non-consensual use of brain-reading technologies, in order to foresee how these cases would be resolved in the light of this majority interpretation.
B. Three Currently Available Brain-Reading Technologies: Functional Magnetic Resonance Imaging, P300 and Semantic Decoding
I. Functional Magnetic Resonance Imaging
Functional Magnetic Resonance Imaging (fMRI) is designed to measure brain activity by measuring blood flow. Because the brain needs oxygen to function, a higher presence of oxygen means a higher level of activity.Footnote 12
In lie detection studies using this technology, people are asked to answer a series of questions while an fMRI scanner monitors their level of brain activity in different areas of the brain.Footnote 13 In some studies, the same question is asked twice and the subject is asked to answer differently each time.Footnote 14 In other studies, the questions are asked only once, but only some of them relate to true information about the crime under investigation, while other questions have a publicly known answer, such as the capital of a country.Footnote 15
The hypothesis behind this technology is that telling the truth is the brain’s natural response, so the level of brain activity should remain within normal parameters. In contrast, lying requires a higher level of brain activity because the person must first remember the truth, then suppress it while creating a lie, and finally express the latter.Footnote 16 To perform this higher level of activity, the brain needs more energy, which it gets from an increased flow of oxygenated blood. These changes can be seen with an fMRI scanner.Footnote 17
Scientists have generally associated the prefrontal region of the brain with the process and action of lying,Footnote 18 at least spontaneously.Footnote 19 Therefore, by identifying the areas of the brain that are activated when a question is answered, it would be possible to reasonably infer whether the answer is a lie or not. In other words, if it is determined that the area of the prefrontal cortex has been activated, by detecting increased blood flow to it, there will be a higher probability that the answer in question is a lie.
II. P300
Event-related potentials are very small voltages generated in the brain structures in response to specific events or stimuli.Footnote 20
The P300 wave is an event-related potential that occurs approximately 300 miliseconds after the presentation of a significant stimulus.Footnote 21
Tests based on the P300 wave, such as the Concealed Information Test (CIT)Footnote 22 or Brain Fingerprinting,Footnote 23 aim to determine whether or not a person recognizes specific information related to a crime by detecting the P300 wave in an electroencephalogram.Footnote 24
In a CIT, the subject is presented with several stimuli divided into three categories: a relevant stimulus, actually related to the crime; irrelevant stimuli, unrelated to the crime in question and belonging to the same category as the relevant stimulus, such that an innocent person would not be able to distinguish them from the object related to the crime; and, finally, a stimulus that has been made relevant by the test instructions, the purpose of which is to secure the subject’s cooperation.Footnote 25 If the presentation of the relevant stimulus elicits a significant physiological response in the subject, it can be concluded that they recognizes the stimulus. As Farwell explains it: “When a subject recognizes and takes note of something significant in the present context, the brain emits an ‘Aha!’ response.”Footnote 26
For example, in a murder case where the victim was stabbed in the heart with a chisel, a chisel would be a relevant stimulus, a knife would be an irrelevant stimulus, and an example of a stimulus made relevant by the test instructions would be a red apple, where the test subject has been instructed to report whenever it sees such an apple.
Although it is possible for an innocent person to respond with a greater physiological response to the relevant stimulus, the possibility of a false positive can be reasonably reduced by increasing the number of questions.Footnote 27
Supporters of tests based on the P300 wave argue that they are not designed to detect lies, but only to determine whether a person recognizes information that is presumably known only to the perpetrator.Footnote 28 This type of test does not seek to determine what information should have been stored in the person’s brain, nor how the information stored in the brain got there.
The test result could indicate that the subject has lied, but this will not always be the case, as the subject can have a plausible explanation for possessing this knowledge, such as a newspaper report.Footnote 29
In addition to helping to investigate offences committed in the past, CIT can also help to prevent future terrorist attacks.Footnote 30
III. Semantic Decoding
The combination of techniques such as fMRI with neurological etiologies has revolutionized our understanding of how semantic concepts are encoded in the brain. This new understanding of how our brains encode semantic concepts has given rise to semantic decoding, a new field of study defined as the decoding of semantic concepts from recordings of our brain activity.Footnote 31
In 2023, researchers created a new artificial intelligence system called “semantic decoder,” which can translate a person’s brain activity while listening to a story or silently imagining telling a story into a continuous stream of text.Footnote 32 This system might help people who are mentally conscious yet unable to physically speak to communicate intelligibly again.Footnote 33
Unlike other language decoding systems, this semantic decoder does not require subjects to have surgical implants, making the process non-invasive. In addition, participants are not restricted to using only words from a prescribed list.Footnote 34
How does the semantic decoder work? First, brain activity is measured using fMRI after extensive training of the decoder, in which the individual listens to 16–non-consecutive–hours of speech input by listening to a series of “naturally spoken narrative stories”:
‘Decoding continuous language thus requires solving an ill-posed inverse problem, as there are many more words to decode than brain images. Our decoder accomplishes this by generating candidate word sequences, scoring the likelihood that each candidate evoked the recorded brain responses and then selecting the best candidate. To compare word sequences to a subject’s brain responses, we used an encoding model that predicts how the subject’s brain responds to natural language. We recorded brain responses while the subject listened to 16 h of naturally spoken narrative stories, yielding over five times more data than the typical language fMRI experiment. We trained the encoding model on this dataset by extracting semantic features that capture the meaning of stimulus phrases and using linear regression to model how the semantic features influence brain responses. Given any word sequence, the encoding model predicts how the subject’s brain would respond when hearing the sequence with considerable accuracy.’Footnote 35
Then, provided that the participant is open to having their thoughts decoded, their listening to a new story or imagining telling a story allows the machine to generate corresponding text from brain activity alone.Footnote 36
It is important to note that the resulting text is not a perfect word-for-word transcript. Instead, the semantic decoder is designed to capture the overall essence of what is said or thought. The decoder produces text that closely—and sometimes exactly—matches the intended meaning of the original words about 50% of the time. For example, in experimental trials, when a person heard the sentence “I don’t have my driver’s license yet,” her thoughts were translated as “She has not even started to learn to drive yet.”Footnote 37
The researchers also addressed concerns about the potential misuse of the semantic decoder. In particular, the decoder has raised questions about the possibility of unauthorized access to people’s thoughts. However, the researchers emphasized that the decoder was only effective with cooperative individuals who willingly participated in the training of the decoder. When tested on untrained individuals or those who actively resisted by thinking other thoughts, the results were incomprehensible and unusable. Nevertheless, the researchers noted that the technology could overcome this limitation in the future and, therefore, pointed out that there are acute reasons to be concerned about the mental privacy issues associated with this new technology.Footnote 38
C. The Distinction Between Testimonial and Real Evidence as a Key Feature of the Interpretation About the Right Against Self-Incrimination: A Comparative Overview
The right against self-incrimination can be understood, in a broad sense, as the right to remain silent and not to contribute to incriminating oneself.Footnote 39
The importance of the right against self-incrimination has been universally recognized. The European Court of Human Rights (ECtHR) has ruled that the right in question is a generally accepted international standard that lies at the heart of the notion of a fair proceeding.Footnote 40 Similarly, the United States Supreme Court has stated that the right against self-incrimination “registers an important advance in the development of our liberty-one of the great landmarks in man’s struggle to make himself civilized.”Footnote 41
Despite its fundamental relevance and the apparent simplicity of its definition, the right against self-incrimination has been characterized as one of the most complex guarantees in the entire body of fundamental rights applicable in the context of criminal proceedings.Footnote 42 Similarly, some authors have affirmed that the self-incrimination clause of the Fifth Amendment to the United States Constitution “is an unsolved riddle of vast proportions, a Gordian knot in the middle of our Bill of Rights.”Footnote 43
Moreover, the development of new technologies has led to other complex scenarios. Can a defendant be compelled to disclose the password to her smartphone? Can a defendant be compelled to decrypt their smartphone by placing their finger on it? Can a defendant be compelled to sign a document authorizing the prosecution to ask national and international banks for their financial records? Can an accused person be forced to sit still while the prosecution carries out a brain fingerprinting test on them in order to detect the presence of information related to the crime under investigation?
According to the traditional interpretation of the right against self-incrimination, its scope of application includes only “testimonial evidence,” thus excluding “real or physical evidence.”Footnote 44 Consequently, for example, the prosecution cannot compel a person to make an incriminating statement, but it can compel him to give a blood sample in order to carry out a scientific examination, or to tolerate the taking of a photograph of his body in order to carry out a recognition procedure.
This interpretation has been adopted by the Supreme Court of the United States, the Supreme Court of Canada, the Constitutional Court of Spain, the Constitutional Court of Chile, and the ECtHR. Of course, majority does not mean right, and minority does not mean wrong. However, the fact that different courts belonging to different legal systems follow a remarkably similar approach illustrates how widespread and well-established this approach is.
I. The United States Supreme Court
The Fifth Amendment to the United States Constitution declares: “No person [. . .] shall be compelled in any criminal case to be a witness against himself [. . .].”
According to the United States Supreme Court, the word “witness” in the Fifth Amendment limits the protection to testimonial evidence.Footnote 45 Therefore, the right in question “applies only to preventing a person from having to give testimony against himself or herself.”Footnote 46
In Holt v. United States, the Supreme Court held that “the prohibition of compelling a man in a criminal court to be witness against himself is a prohibition of the use of physical or moral compulsion to extort communications from him, not an exclusion of his body as evidence when it may be material.”Footnote 47
In Schmerber v. California, the Supreme Court distinguished between testimonial and real evidence. In that case, the defendant, who had been convicted of driving under the influence of alcohol, argued that his right against self-incrimination had been violated when a blood sample was extracted from him at the hospital despite his refusal to consent.Footnote 48
The Supreme Court affirmed the conviction, holding that “the privilege protects an accused only from being compelled to testify against himself, or otherwise provide the State with evidence of a testimonial or communicative nature, and that the withdrawal of blood and use of the analysis in question, in this case, did not involve compulsion to these ends.”Footnote 49 In other words, a compulsion which makes a defendant the source of “real evidence” does not violate the right against self-incrimination.Footnote 50
Since Schmerber, the Supreme Court has decided several cases allowing a defendant to be forced to stand in a lineup,Footnote 51 provide handwriting samples,Footnote 52 give voiceprints,Footnote 53 take sobriety tests that measure mental acuity and physical coordination,Footnote 54 and produce documents.Footnote 55
Notwithstanding, in Fisher v. United States, the Supreme Court recognized that the act of producing evidence in response to a request from the authority has communicative aspects of its own, because compliance with the request tacitly concedes the existence of the evidence demanded and its possession or control by the person in question, as well as the belief that the evidence delivered is the evidence requested.Footnote 56 However, compulsion of these implied admissions alone does not warrant Fifth Amendment protection with respect to the act of production, because the right against self-incrimination “only applies when the accused is compelled to make a testimonial communication that is incriminating.”Footnote 57
Therefore, the Fisher test consists of three elements: (i) Compulsion; (ii) a testimonial communication; and (iii) incrimination.Footnote 58 If any one of the three elements is lacking, the right is unavailable.Footnote 59
Nevertheless, in Fisher the Supreme Court established the “foregone conclusion” doctrine, which holds that however incriminating some documents may be, the act of producing them does not amount to the level of testimony within the protection of the Fifth Amendment where the existence and location of the papers are a foregone conclusion. In that case, the defendant adds little or nothing to the sum of the government’s information by conceding that she in fact has the documents.Footnote 60 Under those circumstances, the act of production does not amount to testimony, but to surrender.
The “foregone conclusion” doctrine was reaffirmed in United States v. Hubbel. In that case, the prosecution obtained a broad-reaching subpoena for production of evidence—duces tecum—from the grand jury investigating the defendant’s criminal misconduct. After the prosecution granted the defendant use immunity, he produced more than 13,000 pages of documents. The contents of the documents provided the prosecution with the information that led to a second prosecution against him for tax offences and mail and wire fraud.Footnote 61
At the outset, the Supreme Court recalled that the word “witness” in the constitutional text limits the relevant category of compelled evidence to those that are “testimonial” in character.Footnote 62
The Court then applied the reasoning of Fisher to the facts of the case. Regarding the “foregone conclusion” doctrine, the Supreme Court held that the prosecution had not shown any “prior knowledge of either the existence or the whereabouts of the 13,120 pages of documents ultimately produced.”Footnote 63 Therefore, the Court concluded that the defendant’s act of producing those documents had been testimonial in nature, and when the prosecution used the documents against him the granted immunity had been violated.Footnote 64
Trying to explain the new doctrine, Pinto has stated that disclosure will be deemed to be a foregone conclusion if the government can demonstrate knowledge of (i) the existence of the evidence demanded; (ii) the possession of and control over the evidence in question by the defendant; and (iii) the defendant´s belief that the documents are those requested by the authority—authenticity. If the government can satisfy all three elements, then compelling the production of such evidence does not violate the right against self-incrimination.Footnote 65
II. The Canadian Supreme Court
Section 11(c) of the Canadian Charter of Fundamental Rights provides that any person charged with an offence has the right not to be compelled to be a witness in proceedings against him or her in respect of that offence.
This provision should be read in conjunction with Section 13, which states that any witness who testifies in any proceeding has the right not to have any incriminating evidence so given used to incriminate him or her in any other proceeding, except in a prosecution for perjury or giving contradictory evidence. Moreover, the Supreme Court has ruled that Article 7 of the Canadian Charter, according to which everyone has the right to life, liberty, and security of the person and the right not to be deprived thereof except in accordance with the principles of fundamental justice, also provides a residual protection against improper compelled testimony, because it is well-established that the principles of fundamental justice include the right against self-incrimination.Footnote 66
In order to determine whether the right against self-incrimination has been infringed, relevant factors include the presence or absence of: (1) Real coercion by the state in obtaining the statements; (2) an adversarial relationship between the accused and the state at the time the statements were obtained; (3) an increased risk of unreliable confessions; and (4) an increased risk of abuses of power by the state.Footnote 67
Regarding the scope of application of the right in question, the Supreme Court has held that Section 11(c) protects the accused from oral or testimonial compulsion.Footnote 68 Consequently, compelling a defendant to give fingerprintsFootnote 69 or breathalyzer evidence,Footnote 70 or the introduction into evidence of the accused’s private journals,Footnote 71 does not engage Section 11(c).
III. The Spanish Constitutional Court and the Chilean Constitutional Court
The Spanish Constitutional Court has also interpreted the right against self-incrimination from a testimonial perspective, stating that it should be interpreted as the right not to testify against oneself and not to confess guilt.Footnote 72
According to the Constitutional Court, the essential content of these rights is the prohibition of compulsion to testify against oneself and the recognition of the freedom to testify or not to testify, and to do so as one thinks best.Footnote 73
Based on this interpretation, the Spanish Constitutional Court has held that the obligation to submit to a breathalyzer test did not violate the right against self-incrimination, because the defendant was not compelled to make a self-incriminating statement.Footnote 74 In another case, the Spanish Constitutional Court ruled that an obligation on a taxpayer to produce or submit accounting documents could not be considered a violation of the right against self-incrimination because such behavior did not amount to an expression of will.Footnote 75
In the same vein, the Chilean Constitutional Court has also interpreted the right against self-incrimination from a testimonial perspective.Footnote 76
The Chilean Constitutional Court has held that the scope of application of the right against self-incrimination under the Chilean Constitution is quite narrow, with the only scope defined in Article 19 No. 7 (f) of the Constitution, which prohibits compelling the defendant in a criminal case to testify under oath.Footnote 77
For the above reasons, the Constitutional Court has held that the criminal offence of unjustified refusal to submit to a test for alcohol or narcotics, provided for and punished by Article 195 bis of the Traffic Law, does not violate the right against self-incrimination,Footnote 78 because it simply does not compel the defendant to make a self-incriminating statement.Footnote 79
IV. The European Court of Human Rights and the Court of Justice of the European Union
Currently, the right against self-incrimination is explicitly recognized in Article 7 of EU Directive 2016/343, which requires EU Member States to guarantee the right to remain silent and not to incriminate oneself. However, this was not always the case.
In contrast to other international instruments that explicitly recognize the right against self-incrimination, such as the International Covenant on Civil and Political RightsFootnote 80 and the American Convention on Human Rights,Footnote 81 the European Convention on Human Rights does not explicitly recognize the right against self-incrimination.Footnote 82
The ECtHR recognized the existence of the right against self-incrimination in Funke v. France, decided in 1993.Footnote 83 In Funke, the ECtHR did not address the rationale of the right against self-incrimination or its scope of application.Footnote 84
Three years later, in John Murray v. United Kingdom, the ECtHR attempted to provide a fuller explanation of its reasons for incorporating the right against self-incrimination into the right to a fair trial.Footnote 85 In that case, the ECtHR characterized the right against self-incrimination as a recognized international standard that lies at the heart of the notion of a fair trial. Therefore, it must be understood that the right against self-incrimination is part of the general right to a fair trial provided for in Article 6 of the European Convention.Footnote 86 This is the current position of the ECtHR.Footnote 87
The Court of Justice of the European Union (CJEU) has ruled along the same lines. The right against self-incrimination first appeared in EU competition law in the 1989 case Orkem v. Commission. In that case, the defendant challenged a Commission decision requesting information, arguing that it infringed upon her right against self-incrimination. The CJEU noted that even though such a right was not included in the European Convention on Human Rights, the Community law prevented the Commission from undermining the right of defense of the defendant.Footnote 88 Therefore, the CJEU ruled that the Commission “may not compel an undertaking to provide it with answers which might involve an admission on its part of the existence of an infringement which it is incumbent on the Commission to prove.”Footnote 89
Over the years, the position of the CJEU has become increasingly similar to that of the ECtHR. In Consob, the CJEU deliberately aligned its interpretation of Articles 47 and 48 of the European Union Charter of Fundamental Rights with the case law of the ECtHR on Article 6. First, the CJEU reaffirmed that Articles 47 and 48 of the Charter, which recognize the right to a fair trial and other procedural safeguards, are equivalent to Article 6 of the European Convention on Human Rights, which provides for the right to a fair trial. Second, the CJEU noted that although Article 6 of the European Convention does not refer to the right against self-incrimination, it is “a generally recognized international standard which lies at the heart of the notion of a fair trial” and has long been recognized like that by the ECtHR. Given that, according to Article 52 of the European Union Charter, the rights set forth therein shall have the same meaning as the rights set forth in the European Convention on Human Rights, the CJEU held that Articles 47 and 48 of the Charter must be construed as including the right against self-incrimination.Footnote 90
In K.B. and F.S., the CJEU reiterated that the right against self-incrimination is safeguarded not only by Article 48 of the European Charter, which recognizes the presumption of innocence and right of defense, but also by the second paragraph of Article 47, which concerns the right to a fair hearing.Footnote 91
With regard to the scope of the right in question, the ECtHR first addressed the issue of material evidence and the right against self-incrimination in the case of Funke v. France. In that case, the customs authority issued an order requiring the defendant to produce bank statements from his foreign accounts. The defendant refused to produce such documents and was subsequently sanctioned for failure to do so. The penalty was a fine, which increased for each day that he continued to refuse to produce the evidence.Footnote 92
The ECHR noted that the customs authority had sanctioned the defendant in order to obtain certain incriminating documents which it believed to exist, although it was not sure that they did. Unable to obtain them by other means, the authority sought to compel the defendant to hand over the documents. The ECtHR held that such use of coercion was contrary to the right against self-incrimination and found a violation of Article 6.Footnote 93
In Saunders v. United Kingdom, the ECtHR drew a distinction between the compelled production of testimonial and real evidence.Footnote 94 The ECtHR held that the right against self-incrimination is primarily concerned with respecting the will of an accused person to remain silent. Therefore, it does not extend to evidence “which may be obtained from the accused through the use of compulsory powers but which has an existence independent of the will of the suspect such as, inter alia, documents acquired pursuant to a warrant, breath, blood and urine samples and bodily tissue for the purpose of -DNA- testing.”Footnote 95
The distinction drawn in Saunders might seem clear. In fact, it was literally transposed into Article 7(3) of Directive EU 2016/343,Footnote 96 according to which the exercise of the right against self-incrimination shall not prevent the authorities from gathering evidence that can be lawfully obtained through coercive methods and that has an existence independently of the will of the persons charged.Footnote 97
In Consob, the CJEU emphasized the testimonial aspect of the right against self-incrimination, stating that since the:
[P]rotection of the right to silence is intended to ensure that, in criminal proceedings, the prosecution establishes its case without resorting to evidence obtained through methods of coercion or oppression in defiance of the will of the accused [. . .], this right is infringed, inter alia, where a suspect is obliged to testify under threat of sanctions and either testifies in consequence or is sanctioned for refusing to testify.Footnote 98
The latest case in which the ECtHR addressed the problematic relationship between the right against self-incrimination and the compelled production of documents was de Legé v. the Netherlands. In that case, the ECtHR resolved the complaint of a person who had been compelled by the Dutch Tax Authority to submit evidence relating to his finances. The evidence provided by the person was used by the Dutch Tax Authority to set tax fines imposed on him.Footnote 99
The ECtHR found no violation of the right against self-incrimination, holding that the use of the evidence produced by the applicant to the authorities did not fall within the scope of the right against self-incrimination because the evidence produced concerned pre-existing documents—therefore not created as a result of the compulsion exerted by the authorities—of whose existence the authorities were already aware.Footnote 100
By deciding the case in that way, the ECtHR adopted a line of reasoning similar to the “foregone conclusion” doctrine of the United States Supreme Court described above.Footnote 101
D. Given the Distinction Between Testimonial and Real Evidence, Does Coercive Brain-Reading Violate the Right against Self-Incrimination?
As stated above, the purpose of this Article is to analyze whether the collection of incriminating information through the non-consensual use of brain-reading technologies violates the right against self-incrimination under its traditional interpretation, according to which the scope of application of this right includes only “testimonial evidence,” thus excluding “real or physical evidence.”
Given that the traditional interpretation of the right against self-incrimination includes only “testimonial evidence,” the answer to the question of whether the collection of incriminating information through the non-consensual use of brain-reading technologies violates the right in question depends on whether the incriminating information obtained through these technologies should be qualified as testimonial evidence. If so, the collection of such information would violate the right against self-incrimination. If not, the evidence would not violate that right.
When should evidence be considered as testimonial? What does it mean for evidence to be testimonial? Unfortunately, distinguishing between testimonial and real evidence is not an easy task, mainly due to the lack of a common concept of “testimony.” In fact, it is possible to identify several different existing concepts.Footnote 102 This is a problem, because if there is no clarity on the concept of testimonial evidence, the application of the traditional interpretation of the right against self-incrimination will not be easy. The same United States Supreme Court recognized as early as 1966 that there would be many cases in which the distinction between testimonial and real evidence would not be easy to make because some “tests seemingly directed to obtain ‘physical evidence,’ for example, lie detector tests measuring changes in body function during interrogation, may actually be directed to eliciting responses which are essentially testimonial.”Footnote 103
The analysis of the case law of the ECtHR shows that the application of the distinction drawn in Saunders has not been entirely consistent, as evidenced by J.B. v. Switzerland and Chambaz v. Switzerland, two cases decided after Saunders.
In J.B. v. Switzerland, the tax authority sanctioned the defendant in an administrative procedure for refusing to hand over documents and information relating to his income. The applicant claimed that the imposition of such sanctions violated his right against self-incrimination.Footnote 104 Regarding the defendant’s argument, the ECtHR found that the authorities were attempting to compel him to give evidence against himself.Footnote 105 The ECtHR concluded that the authorities had indeed compelled the defendant to incriminate himself, and therefore found a violation of the right against self-incrimination.Footnote 106 It must be emphasized that the ECtHR explicitly affirmed that the present case did not involve evidence whose existence was independent of the suspect’s will.Footnote 107
A decade later, the ECHR issued its decision in Chambaz v. Switzerland. In that case, the tax authority assessed the applicant’s taxable income for the year 1989–1990 and found that he had not declared all of his income because the growth of his assets was disproportionate to his stated income. During the proceeding, the applicant was asked to produce evidence, but he refused. As a result, the tax authority sanctioned him. The ECtHR found that, by imposing a fine on the applicant for refusing to provide all the items requested, the authorities had compelled him to produce documents which would have provided information on his income and assets for tax assessment purposes, thereby forcing him to incriminate himself. Once again, the ECtHR found a violation of the right against self-incrimination in a case of compelled production of documentary evidence.Footnote 108
Was the existence of the evidence produced by the defendants in J.B. and Chambaz “independent of the will of the suspect”? Because both cases involved financial documents, one might think that their existence was independent of the defendant’s will. However, the fact that the ECtHR found a violation of the right against self-incrimination points in the opposite direction.
Some legal scholars have proposed to read the Saunders exception in the sense that the right against self-incrimination would prohibit compelling a person to actively cooperate with an investigation against herself.Footnote 109 For example, Trechsel has suggested that the right against self-incrimination “only covers assistance from the suspect which could not be substituted by employing direct force.”Footnote 110 In this view, because bodily samples can be obtained without the active cooperation of the defendant—for example, by using force to take them—that would not violate the right against self-incrimination.Footnote 111 The same is true when the government, by means of an entry and search warrant, seizes documents or evidence from the home of the defendant, from whom active cooperation is not required. On the contrary, when the government orders a person to hand over documentary evidence in their possession, threatening them with the imposition of sanctions if they refuse to comply, the government is demanding active cooperation from the defendant in the investigation, which is contrary to the right against self-incrimination.Footnote 112
This reading of the Saunders exception, which suggests that the right against self-incrimination prohibits compelling a person to cooperate actively in an investigation against herself, makes it possible to understand the judgments in J.B. and Chambaz. However, this is not the case in two other ECtHR decisions: Allen v. the United Kingdom and Van Weerelt v. the Netherlands.Footnote 113 In these two post-Saunders cases, the ECtHR found no violation of the right against self-incrimination, even though the tax authorities compelled the applicant to produce documents relating to his assets and business. In Allen, the ECtHR held that the right against self-incrimination:
[D]oes not per se prohibit the use of compulsory powers to require persons to provide information about their financial or company affairs [. . .] The obligation to make disclosure of income and capital for the purposes of the calculation and assessment of tax is indeed a common feature of the taxation systems of Contracting States and it would be difficult to envisage them functioning effectively without it.Footnote 114
The same reasoning was followed in Van Weerelt.Footnote 115
The ECtHR even explicitly recognized in Van Weerelt that the tax authority had compelled the applicant “to give information that could not be obtained from any other source than the applicant himself for the purpose of levying taxes and interest in accordance with the applicable tax legislation.”Footnote 116 This is of paramount importance because it means that the ECtHR recognized that the tax authority would not have been able to obtain the information sought without the applicant’s cooperation.Footnote 117 In both Allen and Van Weerelt, the fact that taxes were involved seems to have been the decisive argument for dismissing the application.
Considering the decisions in J.B., Chambaz, Allen, and Van Weerelt, it can be stated that the scope and application of the distinction drawn in Saunders between evidence whose existence is independent of the suspect’s will and that which is not is unclear.
In an attempt to provide some clarity, the United States Supreme Court, in Doe v. United States, offered a concept of testimonial evidence, holding that the policies underlying the right against self-incriminationFootnote 118 “are served when the privilege is asserted to spare the accused from having to reveal, directly or indirectly, his knowledge of facts relating him to the offense or from having to share his thoughts and beliefs with the Government.”Footnote 119
In that case, the defendant was asked to produce records of transactions in accounts at three foreign banks. The defendant produced some bank records and testified that no other records were in his possession. When asked about the existence of additional records, the defendant refused to answer, invoking his right against self-incrimination. The government asked the court to order the defendant, without admitting the existence of any account, to sign a consent form authorizing the banks to disclose records of all of his accounts.Footnote 120
Applying the above concept of testimony, the Supreme Court held that the defendant’s execution of the consent form did not have testimonial significance:
[B]ecause neither the form, nor its execution, communicates any factual assertions, implicit or explicit, or conveys any information to the Government. The consent directive itself is not ‘testimonial.’ It is carefully drafted not to make reference to a specific account, but only to speak in the hypothetical. Thus, the form does not acknowledge that an account in a foreign financial institution is in existence or that it is controlled by petitioner. Nor does the form indicate whether documents or any other information relating to petitioner are present at the foreign bank, assuming that such an account does exist.Footnote 121
If a court were to apply the concept of testimonial evidence from Doe v. United States to cases involving the collection of incriminating information through the non-consensual use of brain-reading technologies, the conclusion would likely be that the information obtained through fMRI and P300 technologies should be classified as non-testimonial or real evidence, because the relevant information obtained in these cases is not the communicative content expressed, but the brain function of the defendant: Which areas of the brain the blood is flowing to, if it is an fMRI, or whether or not it is possible to detect a P300 wave, if it is a CIT.Footnote 122 On the contrary, the information obtained through a semantic decoder would constitute testimonial evidence, because in this case the relevant information is indeed the content of what the person in question expresses.
The above reasoning would find support in United States v. Dionisio. In that case, a grand jury subpoenaed approximately twenty persons, including the defendant, to provide voice samples for identification purposes. The defendant refused to comply on Fourth and Fifth Amendment grounds. The district court rejected both claims and held the defendant in contempt.Footnote 123 The United States Supreme Court held that the claim that compelled production of the voice samples would violate the Fifth Amendment was correctly rejected because the “voice recordings were to be used solely to measure the physical properties of the witnesses’ voices, not for the testimonial or communicative content of what was to be said.”Footnote 124
However, the picture is not so simple. In Pennsylvania v. Muniz, the police asked the defendant the date of his sixth birthday to verify his sobriety, a question he was unable to answer correctly. The government argued that this incriminating inference did not trigger the protection of the Fifth Amendment privilege because the inference concerned the physiological functioning of the defendant’s brain, which should be considered “real or physical evidence,” not testimonial.Footnote 125 The Supreme Court held that the above characterization addressed the wrong question because it analyzed the fact to be inferred rather than the defendant’s act. According to the Court, “the question is not whether a suspect’s ‘impaired mental capacity’ can fairly be characterized as an aspect of his physiology, but rather whether Muniz’s response to the sixth birthday question, which gave rise to an inference of such impairment, was testimonial in nature.”Footnote 126
In light of the foregoing, the Supreme Court rejected the argument that the relevant information was only the physical impossibility of answering the question and not the content of the answer. On the contrary, the Court held that the sixth birthday question in this case required a testimonial answer:
The content of his truthful answer supported an inference that his mental faculties were impaired, because his assertion (he did not know the date of his sixth birthday) was different from the assertion (he knew the date was (correct date)) that the trier of fact might reasonably have expected a lucid person to provide. Hence, the incriminating inference of impaired mental faculties stemmed, not just from the fact that Muniz slurred his response, but also from a testimonial aspect of that response.Footnote 127
It could be argued that there is a contradiction between the decisions in Muniz and Dionisio, because in the latter case the defendant also had to perform a testimonial act—speaking—but the Supreme Court ruled out a violation of the right against self-incrimination. Perhaps for this reason, some scholars have been critical of the Dionisio decision on the grounds that it was really a case of testimonial evidence.Footnote 128
Applying now the reasoning in Muniz rather than the approach in Doe, it could be argued that obtaining incriminating information through the non-consensual use of brain-reading technologies that require the subject to perform testimonial acts, such as answering questions, would violate the right against self-incrimination. Conversely, technologies that do not require the subject to perform any testimonial acts, such as the semantic decoder, would not violate this right.
Returning to the question posed at the beginning of this Article, does the collection of incriminating information through the non-consensual use of brain-reading technologies violate the right against self-incrimination according to its traditional interpretation? In my opinion, the review of the case law of the ECtHR and the United States Supreme Court shows that this question is not easy to answer. It will depend on the approach adopted and the concept of testimony advocated, whether the act of the accused or the information obtained should be qualified as testimonial or not. From this perspective, the clarity and certainty that the traditional interpretation of the right against self-incrimination can offer certainly does not seem much.
There is no doubt that technological progress both raises questions that were difficult to imagine until recently and challenges our dogmatic structures and traditional interpretations. Nevertheless, the legal system has the fundamental task of resolving the new problems with coherence and predictability. In view of the difficulties in applying the traditional interpretation of the right against self-incrimination and the limited certainty it provides, it could be time to critically discuss this traditional interpretation, particularly because of its inability to address the problems raised by the development of new technologies.
Although developing an alternative interpretation is beyond the scope of this Article, I believe that such an interpretation should focus on whether the defendant was compelled to cooperate rather than on the nature of the evidence. In fact, such approaches already exist.
For example, in the context of the United States, Nagareda has proposed to abandon the distinction between testimonial and real evidence for the purposes of the right against self-incrimination. Instead, the right in question should be interpreted as a prohibition for the authority to compel the accused to provide evidence, regardless of its nature. In the author’s opinion, the expression “to be a witness” of the Fifth Amendment should be understood as “to give evidence.”Footnote 129 Along the same lines, Lin has proposed to understand that the “Self-Incrimination Clause bars the admission of compelled evidence, testimonial or otherwise. At the time the Fifth Amendment was ratified, there was no semantic difference between “being a witness” and “giving evidence,” and no such difference existed before that time either.”Footnote 130
In a concurring opinion written in United States v. Hubbell, Justices Thomas and Scalia supported this interpretation, stating that there is:
[S]ubstantial support for the view that the term "witness" meant a person who gives or furnishes evidence, a broader meaning than that which our case law currently ascribes to the term. If this is so, a person who responds to a subpoena duces tecum would be just as much a "witness" as a person who responds to a subpoena ad testificandum.Footnote 131
In the same vein, Justice Marshall complained in 1972 that he could not “accept the notion that the Government can compel a man to cooperate affirmatively in securing incriminating evidence when that evidence could not be obtained without the cooperation of the suspect.”Footnote 132
The interpretation described above is analogous to the alternative reading of the Saunder exception, according to which the right against self-incrimination would prohibit compelling a person to actively cooperate with an investigation against herself.
The proposed interpretation is also similar to that developed by German case law, according to which the right against self-incrimination prohibits compelling the defendant to actively cooperate with the prosecution.Footnote 133 For this reason, the scope of the right against self-incrimination in Germany has been extended from oral statements to other acts that require the defendant to perform an activity,Footnote 134 such as producing incriminating documents, actively participating in a psychological or psychiatric examination, providing a voice or handwriting sample, or providing a urine sample. Even exhaling has been considered an “activity,” meaning that no one can be compelled to provide a breath sample for analysis.Footnote 135 However, those forms of evidence gathering that are deemed “passive” on the part of the defendant are therefore permissible even without her consent. As a result, a defendant can be physically forced to submit to the extraction of a blood sample or be held in place so that a search of her body can be undertaken.Footnote 136
If a court were now to apply this alternative interpretation to cases involving the collection of incriminating information through the non-consensual use of fMRI and P300 technologies, as well as through a semantic decoder, the conclusion would likely be a violation of the right against self-incrimination, because all of these technologies require the person to cooperate. For example, fMRI requires the person to answer some questions; P300 technologies, such as CIT or Brain Fingerprinting, need the person to engage with the three different stimuli; and the semantic decoder developed in 2023 is only effective with cooperative people who have willingly participated in the training of the decoder.
E. Conclusions
The development of brain-reading technologies has raised concerns that the authorities could use them to read the minds of defendants in criminal proceedings and obtain incriminating information, posing questions about the right against self-incrimination.
The purpose of this Article was to analyze whether the collection of incriminating information through the non-consensual use of brain-reading technologies violates the right against self-incrimination under its traditional interpretation, according to which the scope of application of this right includes only “testimonial evidence,” thus excluding “real or physical evidence.” Therefore, for example, the prosecution cannot compel a person to make an incriminating statement, but it can compel him to give a blood sample for scientific testing. The Supreme Court of the United States, the Canadian Supreme Court, the Constitutional Court of Spain, the Constitutional Court of Chile, and the European Court of Human Rights have all ruled along these lines.
As a result, the answer to the question of whether the collection of incriminating information through the non-consensual use of brain-reading technologies violates the right against self-incrimination depends on whether the incriminating information obtained through these technologies should be qualified as testimonial evidence.
Unfortunately, a review of the case law of the United States Supreme Court and the ECtHR shows that distinguishing between testimonial and real evidence is not an easy task, mainly due to the lack of a common concept of "testimony." It is also not clear what should be considered testimonial, whether it is the information obtained by the authority or the act that the person concerned is required to perform. From this perspective, the clarity and certainty that the traditional interpretation of the right against self-incrimination can offer certainly does not seem much.
Given the difficulties in applying the traditional interpretation of the right against self-incrimination and the limited certainty it provides, it may be time to abandon this traditional interpretation and develop a different one that focuses on whether the defendant was compelled to cooperate rather than on the nature of the evidence. If the right against self-incrimination were understood to prohibit compelling a person to cooperate in an investigation against himself or herself, and if such an interpretation were applied by courts to cases involving the collection of incriminating information through the non-consensual use of fMRI and P300 technologies, as well as a semantic decoder, the conclusion would likely be a violation of the right against self-incrimination, because all of these technologies require the person’s cooperation.
Acknowledgements
I would like to acknowledge the Alexander von Humboldt Foundation for supporting the completion of this work with a “Humboldt Research Fellowship.”
Funding Statement
No specific funding has been declared in association with this Article.
Competing Interests
The author declares none.
Open access
