When a Reviewer Asks for Experiments You Cannot Do.

When a reviewer asks for an experiment you cannot run, you do not have to choose between doing the impossible and withdrawing the paper. Sort the request into one of three buckets: essential to the paper's claim, desirable but not load-bearing, or outside the study's scope. For anything that is not essential, editors routinely accept a reasoned decline paired with an alternative, such as a fresh analysis of existing data, an added limitation, or a narrower claim. The refusal that works names why the experiment is out of reach and what you did instead.

Grad students rarely control the budget, the equipment, or the timeline a reviewer assumes you have. A reviewer at a well-funded lab may picture a facility you do not share. So the first job is triage, not argument.

How do I decide which requested experiments actually matter?

Sort each request by what it does to your central claim. If the paper's conclusion collapses without the experiment, it is essential. If the experiment would strengthen the paper but the conclusion stands without it, it is desirable. If it tests a different question, it is out of scope.

Read the comment and ask one thing: does my main result survive if I never run this? A request to replicate your key finding in a second cell line, when your whole claim rests on that finding, sits in the essential bucket, and no amount of framing moves it out. A request to add a third dose group to a curve that already shows the effect is usually desirable. A request to extend your survey study with an animal model tests a question your paper never asked, and that one is out of scope.

Do this triage before you write a single word of response. The bucket decides your strategy: essential means find a way or rescope the claim, desirable means offer a substitute, out of scope means make the boundary argument. The gap between a major and a minor revision matters here too, since a major revision gives the editor room to accept substitutes and signals the paper is worth saving.

Can I say an experiment is beyond the scope of the paper?

Yes. "Beyond the scope" is a legitimate argument when the requested work tests a different question, not when it would simply be inconvenient. Editors accept it when you define the scope you actually claimed and show the request falls outside it.

The argument fails when it is a shield. If you write "this is beyond the scope of the present study" about the one control that would confirm your mechanism, the editor reads it as evasion, and so does the reviewer. Scope is a boundary you drew in the introduction, not one you invent in the rebuttal.

Make the boundary concrete. Point to the research question as you stated it, show that the requested experiment answers a different one, and where you can, name that experiment as a good direction for future work in the discussion. That last move earns its place: it tells the reviewer their idea was read, not brushed aside. A scope argument that ends with a sincere future-work line lands differently from a flat no.

What are the alternatives to running a new experiment?

Three substitutes carry weight with editors: a new analysis of data you already hold, an added limitation that concedes the gap honestly, and a narrower claim that no longer needs the experiment. Offer the strongest one the request allows.

Start with reanalysis. Reviewers often ask for new data when a new cut of existing data answers the same worry. If they doubt a confound, a subgroup analysis or a sensitivity check may settle it without a new sample. This is the cheapest substitute and often the most persuasive, because it engages the concern behind the request.

When you cannot close the gap, name it. A limitation written in your own words costs you less than you fear, and it protects the paper from the charge of overclaiming. Pair the limitation with a matching trim to your conclusion so the claim and the evidence line up.

One caution about a tempting shortcut. If you plan to lean on your own unpublished results to stand in for the requested experiment, check the target journal's reference policy first. PLOS ONE, for instance, is explicit: "Do not cite unavailable and unpublished work, personal communications, or rely upon retracted research in the reference list." A pointer to data the editor cannot see does not count as evidence, and at some journals it breaks a submission rule.

How do I phrase a refusal that keeps the door open?

State the decline in one plain sentence, give the reason the editor can check, then show what you did instead. Concede what is true in the comment before you contest the prescription.

The shape borrows from any point-by-point response: quote the comment, agree with the observation inside it, say what you did not do, and explain why in terms of cost, time, or scope the editor can weigh. Naming a real constraint is not an excuse. A biosafety approval that takes eight months, a reagent no longer sold, a clinical cohort that has closed: these are facts an editor can accept. Vagueness is what reads as dodging.

Keep the tone level. The rules that govern disagreeing with a reviewer politely apply doubly to a refusal: no defensiveness, no adjectives standing in for evidence, no wounded pride. For the full frame of the reply, the response to reviewers letter sets it out.

What if the reviewer is right and you still cannot run it?

Sometimes the reviewer is right, the experiment is essential, and it is out of reach. Then the honest options are to rescope the paper to a claim your data support, or to move it to a journal whose bar your current evidence clears.

This is the least welcome answer and the one that saves the most time. If your central claim depends on an experiment you cannot run, no rebuttal phrasing rescues it, and pushing back only delays a rejection. Cut the claim down to what you can defend, or take the paper to a venue where the present evidence is enough. Reviewers who feel dismissed can find you again on the next submission, so a paper narrowed with care travels better than one defended past its evidence. Waraq's manuscript review flags where a claim outruns the data before an editor does, which is the point where this decision costs you the least.