Reading a Matter Reviewer Report

The Matter Reviewer takes a document, or several, and returns a written report on it. It is the tool people underuse, usually because they assume it is a proofreader. It is not. It reads the filing the way a careful senior would read it before it goes out, and the value is in the reading rather than in the checking.

This is a walkthrough of running one and, more importantly, of reading what comes back.

Two modes, and they are not settings

The first choice on the page is the one that matters most, and it is not a preference. It changes whose document the reviewer thinks it is holding.

Matter mode reads the filing as yours. It looks for what is missing, what is inconsistent with the rest of the record, what relief you have pleaded facts for but not asked for, and what a court is likely to raise. It is the review you want before a filing leaves your chamber.

Adverse mode reads the document as the other side's. The reviewer is told, explicitly, that the material belongs to the opposing party, and it reads for the weaknesses you would want to be arguing against. It is the review you want when a written statement or a counter affidavit lands and you have to answer it.

The two modes finding different things is the tool working correctly rather than a defect. The same document read as your own and read as an opponent's should produce two different reports, and if it did not, one of the two readings would be worthless.

Running one

Give the reviewer a document, or several, or paste the matter text. It takes uploads and pasted text together, so a file plus a few paragraphs of context about what stage the matter is at is a perfectly normal submission, and the context genuinely changes the reading.

Set your jurisdiction before you submit. The review is conducted against the law of the jurisdiction selected in the switcher, so a Sindh agreement reviewed with Punjab selected is being read against the wrong statute book.

The confirmation step before submission asks whether the documents belong to a single matter. Answer it honestly. The reviewer looks for inconsistency between documents, and inconsistency between two unrelated files is not a finding, it is noise.

One run is one instrument, whatever happens internally.

Reading the report

The report is a letter rather than a dashboard. It opens in the first person, in the register of an advocate writing to another advocate, and it runs in four numbered sections.

Section I describes the document. Plain factual account of what the reviewer understood it to be: the instrument, the parties, the forum, the stage. Read this first and read it properly. If the reviewer has misidentified the document, everything after it is answering a different question, and you have learned that in ten seconds rather than at the end.

The findings are ranked, and the ranking is real. They come back ordered from most to least significant, with genuine severities rather than a flat list where everything is equally important. The first three findings are the reviewer's actual argument about what is wrong with the document. Treat the tail as a checklist and the head as an opinion.

Each finding carries its own headline and its trail. The headline is the claim in one line. The trail is what in the material led to it, so you can go back to the page and disagree.

The demotion, and why you should notice it

One behaviour is worth understanding because it changes how much weight to put on a finding.

The reviewer holds the names and numbers of statutory provisions rather than their full text. That means there is a specific kind of claim it is not equipped to make: that a section positively requires something. When a finding rests on a statutory requirement, the reviewer has to be able to quote the basis for it. Where it can, the finding stands where its severity puts it.

Where it cannot, the finding is moved down out of the leading position rather than deleted. It is still in the report and still worth your eye, because the instinct that something deserves attention is frequently right even when the account of why is not. What it stops doing is presenting itself as settled law at the top of your reading.

So a finding near the top with a quoted basis is one you can act on. A finding lower down asserting a requirement without one is a prompt to go and look, not a conclusion.

What is never a finding

A closely related rule, and the one that fixed the reviewer's worst habit. Failure to cite something is not by itself a defect. The reviewer used to be handed reference material and grade documents against it as though the list were a compliance standard, which produced reports demanding eighteen mandatory sections in a one page eviction notice and complaining that a will had omitted provisions the governing Act disapplies to that testator. There is no such standard, and the reviewer no longer invents one. A document is reviewed against the law, not against a list of everything that could conceivably have been mentioned.

After the run

Every successful review is saved to your workspace under reviewed documents, with the report alongside it, so the reading survives the session. An adverse review is titled as one so you can tell the two apart six weeks later, which you will need to.

The reviewer's output is an aid to your judgement and not a substitute for it. It is very good at noticing what is absent, which is the thing a human reader is worst at. It is not the one deciding whether the filing goes out.