Burden of Proof: Overview

If the notion of civil rights or civil liberties entails some fundamental freedoms from governmental overreaching, one of the most telling but perhaps subtle expressions of a commitment to the preservation of civil liberties is found in the legal concept of burden of proof. The burden of proof is traditionally described as a procedural device relating to the rules of engagement in the conduct of a trial. It sets out the parameters for which side should win a close case when the evidence is in equipoise or is ambiguous. However, this feature, the effect of resolving close cases, reveals that burdens of proof also have substantive effect; that is, they often represent normative judgments about who should prevail, who is more likely to be ‘‘correct,’’ what substantive policy is to be advanced, and so forth. This aspect of burdens of proof is often obscured by the emphasis on its mechanical or procedural features, particularly when the area under examination is civil. The burden of proof as a commitment of substantive law emerges more clearly in the criminal context.

Burdens of proof are typically described, thanks to Thayer’s 1890 discussion in the Harvard Law Review, as consisting of two parts—the burden of production, that is, who is responsible for coming up with evidence, and the burden of persuasion. Most often the burden of both production and persuasion is on the same party, usually the party bringing the suit, that is, the plaintiff, or in the case of a criminal prosecution, the prosecutor. However, this is not always the case, and sometimes meeting a production burden with respect to a particular issue allows a party to then shift the burden to his opponent, such as the duty to produce counterevidence and/or to persuade the trier of fact as to an alternative interpretation of the evidence presented.

One of the most familiar examples of this shifting of burdens is the self-defense defense to a homicide prosecution. Of course, the prosecutor has the burden of production and persuasion as to the theory of homicide. But this burden can be increased if the defendant pleads and puts on a self-defense case. Typically, the defendant bears the burden of product and some burden of persuasion that the killing with which he is charged was done in self-defense. Assuming that the defendant meets this burden, the burden would then shift back to the prosecutor to disprove the affirmative defense beyond a reasonable doubt. Because the Constitution requires that the prosecution prove its case beyond a reasonable doubt, the concept of an affirmative defense complicates the prosecutor’s burden. The Model Penal Code tends to place the burden on the prosecutor to disprove any affirmative defenses. But because the task of anticipating all affirmative defenses is a daunting one, not too mention somewhat inefficient, this duty is often not triggered until the defense is raised by the defendant with the appropriate amount and quality of evidence—how much evidence the defendant must produce and the nature of his burden, whether preponderance, or some other standard, varies.

In addition, burdens of proof intersect to a large degree with yet another device with both procedural and substantive aspects—the presumption. Presumptions are evidentiary devices which require that a particular fact be assumed to be true or to exist in the absence of evidence to the contrary. Such presumptions can be conclusive or rebuttable, shifting or ‘‘bursting,’’ but they function in much the same way as burdens of proof: to describe the contours of the way the law asserts the world should be or probably is. Thus, often discussions of burdens of proof often also involve the interaction of burdens of proof with presumptions.

A burden of proof of some kind applies to all cases, both civil and criminal. However, the quantum of proof that will satisfy the burden differs depending on whether a case is criminal or civil. And within the civil area the burden of proof can vary and shift depending upon the nature of the cause of action. Typically, however, in a civil case the burden, usually on the plaintiff, is a preponderance of the evidence. In certain types of cases, the burden may be the purportedly higher standard of ‘‘clear and convincing.’’ Arguably, however, the most important and well-known burden of proof, which is relevant for purposes of civil rights and civil liberties, is the burden of proof that distinguishes most criminal cases—beyond a reasonable doubt.

In the area of criminal law, the burden of proof has been an issue of constitutional dimension ever since the Supreme Court’s decision in In re Winship in 1970, although even prior to this decision, ‘‘beyond a reasonable doubt’’ was apparently the governing standard for criminal cases for some time, perhaps since this country’s founding. The ‘‘beyond a reasonable doubt’’ burden applies to both state and federal criminal cases. Thus, with respect to the criminal law, the burden of proof has an explicitly substantive aspect. Academic and political commentators often view the ‘‘beyond a reasonable doubt’’ standard in criminal cases as a forthright acknowledgment that the playing field is not level. Like a handicap in golf, the burden of proof of ‘‘beyond a reasonable doubt’’ arguably represents an attempt to even out some differences in the distribution of advantages as between the state and a defendant. At a minimum, the higher standard recalibrates the array of factors that would normally heavily favor the government.

As a practical matter, the state (whether state or federal) typically has little difficulty in convincing a jury that it has met its burden, at least in the average criminal case. Nevertheless, the common perception is that placing the burden of proof on the state in criminal cases and making that burden a high one, represents a nontrivial limitation on the potential for governmental overreaching. Still, most cases do not go to trial and it can be argued that burdens of proof that apply to the conduct of trials have less impact when it comes to settlement and pleas. Given that the state’s probability of prevailing is, in general, usually perceived to be fairly high, and defendants can be presumed to know that, it may be that the state is often able to use this knowledge to exert pressure on a defendant to enter a guilty plea even when the state might be at some risk if it had to go to trial. That is, the state can still use its power to bluff about its evaluation of the likelihood of prevailing in a particular case. Nevertheless, the burden of proof lurks in the background of any plea negotiations, serving as something of a counterweight, a means by which a defendant can call the state’s bluff in a close case.

Apart from whether placing the burden of proof on the government at trial accomplishes the substantive goals asserted as its justification when, in fact, most cases settle, there is the issue of whether, even if a case goes to trial, the verbal formulation of ‘‘beyond a reasonable doubt’’ accomplishes what it is intended to do. There is some evidence from jury studies that juries do not understand this particular verbal formulation as placing the evidentiary burden on the government that lawmakers assume or intend for it to do. Indeed, such experiences led to the abandonment in the United Kingdom of this formulation in favor of one that requires a jury to be ‘‘firmly convinced’’ of a defendant’s guilt before conviction.

Although the criminal case is the paradigmatic case of the use of a procedural device like the burden of proof to further goals related to civil liberties, its use or impact on civil liberties and civil rights is by no means limited to criminal cases. All manner of civil cases implicate important civil rights or liberties. Just a few of these are civil commitment proceedings, termination of parental rights, legitimacy proceedings, civil forfeitures, dependency proceedings, and other similar actions involving liberty, property, or family issues. In addition, there are civil causes of action that explicitly relate to civil rights such as actions under Section 1983 and Title VII. Adjustments to the burdens of proof, and shifts in understanding of who bears the burden and what sorts of evidence are necessary to meet that burden, have long been ways in which it is possible to track a strand of skepticism, perhaps even fear, in the creation of a variety of causes of action related to civil liberties and civil rights.

From time to time, burdens have been raised and shifted to suggest that claims related to civil rights or their protection have favored status, such as in New York Times v. Sullivan when the Supreme Court raised the burden of proof in defamation against public figures to exclude causes of action with lower burdens of proof. (Actually, the Court changed the substantive law in a way that had the effect of raising the burden of proof, thus providing an excellent illustration of the issue described above of the degree to which burdens of proof implicate both substance and procedure, and presumptions and substantive evidentiary elements. What the Court did, among other things, was to rule that a presumption of damages, once a ‘‘libel per se’’ case had been made, was unconstitutional. One feature of this presumption was to shift the burden to the defendant to prove that his statements were true.)

In other cases, such as Texas Department of Community Affairs v. Burdine, changes in the understanding of who had the burden of proof have arguably signaled the shift to disfavored status for the protection of civil rights. In Burdine, the Supreme Court held that in a Title VII action, if an employer could prove the existence of a nondiscriminatory reason for discharge, the burden of proof shifted back to the employee/plaintiff to show that it was the improper, discriminatory motive that had led to the discharge, not the nondiscriminatory motive.

Tinkering with burdens of proof, either of production or persuasion, represents a way to ‘‘stack the deck’’ in particular kinds of cases. Thus, although facially neutral, burdens of proof can betray current normative positions about the desirability of the pursuit of a particular type of claim or of the likelihood that a particular type of claim is meritorious.


References and Further Reading

  • Allen, Ronald J., Burdens of Proof, Uncertainty, and Ambiguity in Modern Legal Discourse, Harvard Journal of Law and Public Policy 17 (1994): 627–46. 
  • Cleary, Edward W., Presuming and Pleading: An Essay on Juristic Immaturity, Stanford Law Review 12 (1959): 5–28. 
  • Jeffries, Jr., John Calvin, and Paul B. Stephan, III., Defenses, Presumptions, and Burden of Proof in the Criminal Law, Yale Law Journal 88 (1979): 1325–427. 
  • Solan, Lawrence M., Refocusing the Burden of Proof in Criminal Cases: Some Doubt About Reasonable Doubt, Texas Law Review 78 (1999): 105–47. 
  • Thayer, James B., The Burdens of Proof, Harvard Law Review 4 (1890): 45–70. 
  • Underwood, Barbara D., The Thumb on the Scales of Justice: Burdens of Persuasion in Criminal Cases, Yale Law Journal 86 (1977): 1299–348. 

Cases and Statutes Cited

  • In re Winship, 397 U.S. 358 (1970) 
  • Texas Department of Community Affairs v. Burdine, 450 U.S. 248 (1981) 
  • New York Times v. Sullivan, 376 U.S. 254 (1964) 

See also Due Process; In re Winship, 397 U.S. 358 (1970); Proof Beyond a Reasonable Doubt; Title VII and Religious Exemptions


reload, if the code cannot be seen