Making State Law in Federal Court
Benjamin C. Glassman
Abstract
Open-access reader
Benjamin C. Glassman
Abstract
Open-access reader
Abstract: We know from Erie R.R. Co. v. Tompkins that unless the Constitution or a federal statute provides the rule of decision in federal court, state law does. Contrary to the assumption of several recent commentators, however, Erie itself does not tell the federal court how to ascertain what is the law of the state, and the refrain that federal courts are to predict what the state supreme court would decide not only proves unhelpful upon examination, but also has tended to confuse the courts themselves in recent years. Yet federal courts routinely face questions of state law that admit of no clear answer under state precedents.\nIn this Article, I argue that federal courts must exercise their own independent judgment in resolving unsettled state-law issues according to their own calculations of best outcomes, and I explain why this approach comports with the principles of Erie and its progeny. Two other possible positions are flawed: one, which has gained currency in the last decade, is that federal courts are incompetent to opine on unsettled state-law issues and should therefore employ every artifice to avoid doing so, and the other, which derives from academic work of two decades earlier, suggests that federal courts have superior technical competence and should therefore instruct state courts regarding the proper resolution of unsettled state-law issues. My approach also clarifies the meaning of the “prediction of state law” metaphor and answers the troublesome question of the proper weight for district courts to accord federal appellate predictions.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Abstract: We know from Erie R.R. Co. v. Tompkins that unless the Constitution or a federal statute provides the rule of decision in federal court, state law does. Contrary to the assumption of several recent commentators, however, Erie itself does not tell the federal court how to ascertain what is the law of the state, and the refrain that federal courts are to predict what the state supreme court would decide not only proves unhelpful upon examination, but also has tended to confuse the courts themselves in recent years. Yet federal courts routinely face questions of state law that admit of no clear answer under state precedents.\nIn this Article, I argue that federal courts must exercise their own independent judgment in resolving unsettled state-law issues according to their own calculations of best outcomes, and I explain why this approach comports with the principles of Erie and its progeny. Two other possible positions are flawed: one, which has gained currency in the last decade, is that federal courts are incompetent to opine on unsettled state-law issues and should therefore employ every artifice to avoid doing so, and the other, which derives from academic work of two decades earlier, suggests that federal courts have superior technical competence and should therefore instruct state courts regarding the proper resolution of unsettled state-law issues. My approach also clarifies the meaning of the “prediction of state law” metaphor and answers the troublesome question of the proper weight for district courts to accord federal appellate predictions.
Key concepts: Law, State (computer science), Political science, Computer science, Algorithm