Discovery evolved out of an unique feature of early equitable pleading procedure before the English Court of Chancery: among various requirements, a plaintiff's bill in equity was required to plead "positions." These were statements of evidence that the plaintiff assumed to exist in support of his pleading and which he believed lay within the knowledge of the defendant. They strongly resembled modern requests for admissions, in that the defendant was required to plead only whether they were true or false. At some point between the reign of Elizabeth I (1558-1603) and the late seventeenth century, positions were gradually replaced by interrogatories—written questions which the defendant was required to truthfully respond to under oath in his answer to the bill based on information within his own personal knowledge as well as documents in his possession. But back then, interrogatories could only elicit admissible evidence (not the broader modern standard of "reasonably calculated to lead to the discovery of admissible evidence") and could only request evidence in support of the plaintiff's case, not either side's case (that is, they could not ask for evidence which the defendant intended to use in support of his defenses and was otherwise entirely irrelevant to the plaintiff's case). Even worse, this was purely a one-way procedure, because interrogatories could only be pleaded as part of a bill (a pleading initiating a suit in equity). A defendant who needed to obtain evidence in support of his defenses had to file a cross-bill against the plaintiff to plead his own interrogatories.
Discovery did not exist at common law, but its availability in equity attracted litigants in actions at law (legal proceedings in the common law courts). They began to file bills in equity to obtain discovery in aid of actions at law. This led to another innovation in the mid-15th century: the bill to perpetuate testimony of a potential witness. This was for witnesses whose advanced age or poor health implied they would not survive to testify at the trial of an action at law. In this type of proceeding, the parties merely pleaded written interrogatories which a master (in or near London) or a lay commissioner (outside of London) would read out loud to the witness in a closed proceeding without parties or counsel present. The witness's oral answers were written down by the master or lay commissioner in summary form as if they had been delivered as a single continuous narrative (rather than as responses to discrete questions); the actual sequence of questions and answers was not transcribed verbatim like a modern deposition. The resulting document was then sealed and not revealed or "published" (in the terminology of the time) to parties or counsel until shortly before the trial in which it was to be used.
This procedure for ex parte out-of-court pretrial examinations came to be called a "deposition." It continued to be used as an evidence preservation device in aid of actions at law, but it also became the standard method for developing the factual record to be used in courts of equity as derived from the knowledge of third-party witnesses (not merely those who were old or dying). The master or commissioner necessarily performed a limited factfinding function when summarizing testimony in narrative form to be relied upon by the Lord Chancellor in lieu of live testimony in open court. As implied by the secret nature of the proceedings and the absence of parties and counsel, equity's factfinding process was fundamentally inquisitorial (i.e., driven by the court), and not adversarial (i.e., driven by the parties). It is generally believed that this came about because the early Chancellors and the masters who assisted them were clerics with training in Roman and canon law, and therefore had some knowledge of the inquisitorial system as it functioned in ecclesiastical courts. The secrecy of the proceedings was thought to be absolutely essential to prevent perjury and witness tampering; the witnesses would thereby be forced to testify from memory alone, and the parties could not use the facts disclosed in testimony to guide their discovery or litigation strategy. Consistent with this inquisitorial view, there were also prohibitions on repeat testimony and on additional testimony after publication. Rather, the witnesses would testify independently of each other before publication, then at the moment of publication, all would be revealed, and the parties would make their arguments to the Chancellor on that cold record. It is this procedure to which the United States Congress was referring in an 1802 law providing that "in all suits in equity, it shall be in the discretion of the court, upon the request of either party, to order the testimony of the witnesses therein to be taken by depositions."
The next major development (which would remain a unique feature of American and Canadian discovery) occurred under the supervision of Chancellor James Kent of the New York Court of Chancery during the early 19th century. He was trying to respond to the obvious defect of traditional depositions: since parties could not adjust their questions on the fly, they had to propound broadly drawn interrogatories, and in turn elicited "long and complicated accounts" of the facts that were difficult for masters to summarize in writing. Therefore, Kent allowed New York masters to actively engage in oral examination of witnesses (in the sense of formulating questions in real time and narrowing their scope based on the witnesses' answers), and he also allowed parties and counsel to be present when such examinations were conducted. Kent's innovations spread into American federal practice in 1842 when the U.S. Supreme Court amended the Federal Equity Rules to allow masters in equity suits in federal courts to conduct oral examinations of witnesses. However, with the parties and counsel now present to help guide the course of the master's oral examination of the witness, it was inevitable that counsel would insist on taking over the examination itself. Also, their presence meant the proceedings were no longer secret.
Major reforms enacted in New York in the late 1840s and in England in the early 1850s laid the foundation for the rise of modern discovery by imposing a clear separation between pleadings and discovery as distinct phases of procedural law. Discovery devices could now be invoked independently of the pleadings. The New York reforms went much farther, by directly merging common law and equity procedure (which would also happen in England in the early 1870s), and by expressly authorizing pretrial oral examinations of both opposing parties and third-party witnesses, the basis of the modern deposition. (Up to that point, discovery from able-bodied opposing parties was still limited to interrogatories.) In 1861, Rule 67 of the Federal Equity Rules was amended to make deposition by oral examination the regular method of taking evidence in equity; taking witness testimony by written interrogatories was now the exception. Although depositions were still taken in front of court-appointed examiners, their role had been reduced to the preparation of summary narratives to be relied upon as evidence by the court. In 1892, Rule 67 was again amended to require the preparation of an exact transcript. Subsequent amendments in 1893 and 1912 eliminated the deposition's traditional role as an equitable factfinding device by first allowing and then requiring oral testimony in open court in trials of federal suits in equity, thereby reducing the deposition to its modern role in American civil procedure as a discovery and evidence preservation device.