Consecutive or simultaneous interpreting at a deposition

Sworn testimony at a deposition is interpreted consecutively: the question is put, then rendered, then answered, then rendered. Simultaneous interpreting, delivered continuously while someone else is speaking, has a legitimate role at a deposition but not for the testimony itself. The reason is the transcript.

The two modes, precisely

Consecutive interpreting means the speaker delivers a segment, stops, and the interpreter renders it. Each rendering is a separate, complete utterance with a clear beginning and end. Simultaneous interpreting means the interpreter renders continuously while the speaker keeps talking, usually quietly or through equipment, a few seconds behind.

Both are professional skills and neither is a lesser version of the other. Simultaneous is more cognitively demanding to sustain, which is why long simultaneous work is staffed with two interpreters. What separates them for deposition purposes is not difficulty, it is what each produces on the record.

Why sworn testimony is rendered consecutively

A deposition transcript is a sequential record of discrete utterances attributed to identified speakers. Consecutive interpreting produces exactly that: question, rendering, answer, rendering, each one a clean unit the reporter can capture and attribute.

Simultaneous interpreting does not. Two people are speaking at once by design. A reporter cannot cleanly attribute overlapping speech, and the interpretation that ends up in the transcript becomes an approximation of something that was delivered over the top of the original. If the accuracy of a specific answer is ever challenged, there is no discrete rendering to examine.

A national association of court interpreters takes the same view of witness testimony in court. The National Association of Judiciary Interpreters and Translators published a position paper in November 2021 titled "Consecutive Interpreting at the Witness Stand: why it should always be the only mode used." The paper addresses testimony in court under the federal Court Interpreters Act rather than depositions, and argues that interpreting questions simultaneously may distort the witness's understanding of them. The title states the position. It is worth citing when someone treats mode as a scheduling preference.

What simultaneous is legitimately for

There is a real use for simultaneous interpreting at a deposition, and it is not the testimony. It is keeping a non-English-speaking party informed of everything else that happens in the room: colloquy between counsel, objections, discussions about exhibits, scheduling. A party has a real interest in following their own proceeding, and rendering all of that consecutively would add a great deal of time.

So the working arrangement at a well-run deposition is often both. The deponent's testimony is interpreted consecutively on the record. Everything around it is interpreted simultaneously for the party's benefit, off the record. Where that arrangement is used, say so on the record at the start so the transcript reflects what was happening.

What each mode does to the reporter's job

This is the part that gets overlooked, because the reporter is rarely consulted about mode even though they carry the consequence.

• Consecutive gives the reporter clean turn boundaries and one voice at a time. It is more time on the clock and materially less risk in the transcript.
• Simultaneous gives the reporter overlapping speech, uncertain attribution, and a rendering that trails the original by a few seconds, which is the point at which a transcript starts to contain sequencing that did not happen in that order.
• Mixing the modes without stating the switch on the record produces a transcript in which nobody can later tell which passages were rendered discretely and which were rendered over the top.

If you work at a court reporting agency and have ever received a complaint about an interpreted transcript, there is a reasonable chance the underlying cause was mode rather than the interpreter or the reporter.

Where the disputes come from

Interpretation disputes at depositions cluster around a few recurring situations, and mode is upstream of most of them. A question is put, the interpreter renders it, the deponent answers something adjacent, and on review nobody can establish whether the deponent misunderstood the question or the rendering shifted it. With consecutive interpreting each rendering is a discrete utterance that a check interpreter in the room, or anyone reviewing an audio or video recording, can examine; the transcript itself records only the English. With simultaneous there frequently is no discrete rendering to examine.

The related mechanism for putting an interpretation dispute on the record properly is covered in our article on what a deposition interpreter must state on the record, since the credential statement is usually where any challenge begins.

Exhibits sit outside both modes

Handing a deponent a document in a language they cannot read is a third task, and it is neither consecutive nor simultaneous interpreting. It is sight translation: rendering a written document aloud, on the spot, into the spoken target language.

Sight translation is a distinct skill with its own risks, because the interpreter is reading and rendering at once, often with no chance to study the document beforehand. For a short, simple exhibit it is routine. For a dense contract, a medical record, or a document whose exact wording is the point in dispute, the right answer is a written certified translation prepared in advance rather than a rendering improvised in the room.

The practical rule: if the precise wording of a document will matter to the testimony, do not sight translate it. Get it translated beforehand and mark the translation. Deciding this in advance is also what stops a deposition losing half an hour while everyone waits.

Check interpreters and how disputes get raised

Opposing counsel may bring their own interpreter to monitor the rendering. Mode determines whether that monitoring can produce anything usable. With consecutive rendering there is a discrete utterance to compare against, so a check interpreter can point to a specific rendering and state a specific disagreement. With simultaneous there often is not, and the objection collapses into an impression.

The workable protocol is the same one that works remotely: the check interpreter raises the concern to counsel, counsel states it on the record, the interpreters confer, and the agreed rendering goes in. What does not work is two interpreters correcting each other live, which produces a transcript in which the disagreement is preserved but the testimony is not.

Duration, fatigue, and when a second interpreter is needed

Accuracy in either mode degrades with sustained work, which is why long proceedings are staffed with two interpreters working as a team. NAJIT published new guidelines on team interpreting in May 2026, expanding on its 2020 position paper, and they expressly cover depositions rather than only in-court work.

A note of caution about a number you will see quoted: thirty minutes is often given as the switching interval between interpreters. NAJIT's May 2026 guidelines use the figure differently. NAJIT's position is that a team be required for any legal proceeding, in or out of court, that may involve more than 30 minutes of continuous simultaneous interpreting or more than 60 minutes of continuous consecutive interpreting. The switching interval itself is for the team to agree in advance (the guidelines give every 15 to 20 minutes as an example), and 30 minutes is named as the switching standard only for American Sign Language interpreters. Behind the numbers the principle is the same: sustained interpreting degrades and teaming exists to prevent it. Our article on interpreter teams for long proceedings covers the staffing question, and the effect on cost is in what a court interpreter costs in California.

What the deponent should be told beforehand

Consecutive interpreting feels unnatural to someone who has never given testimony through an interpreter, and the resulting behavior causes real problems on the record. A deponent who has not been prepared will answer before the question has been rendered, will keep talking through the rendering, or will speak to the interpreter rather than to the questioner.

A short briefing prevents all three. The deponent should be told to wait for the full interpretation before answering, to answer in short segments rather than long uninterrupted narratives, to speak to counsel rather than to the interpreter, and to say plainly if they do not understand a question rather than guessing at it. None of that coaches the substance of the testimony, and all of it improves the accuracy of the record.

Our article on preparing a non-English-speaking witness for deposition covers the wider preparation question for counsel.

Practical points for scheduling

• Budget the extra time for consecutive rendering at scheduling, not on the day.
• State the mode on the record at the start, including any simultaneous rendering being provided to the party off the record.
• On a remote deposition, state the turn-taking protocol explicitly, because platforms punish overlap harder than a conference room does.
• For a long day, raise teaming when you book rather than when the interpreter is already tired.

Sources

National Association of Judiciary Interpreters and Translators, "Consecutive Interpreting at the Witness Stand: why it should always be the only mode used" (November 2021), and "NAJIT Guidelines on How to Manage and Deliver Team Interpreting Services" (May 2026), both listed at najit.org; California Evidence Code sections 750 and 751. Verified on November 5, 2026.

AMS interpreters render sworn testimony consecutively as standard and will state the mode on the record. See our deposition interpreting service or our trial interpreting service for multi-day proceedings.