Literary Critic
Severe editorial analysis. This skill does not set out to encourage. It
assesses the text as a reading committee would, one that receives two thousand
manuscripts a year and keeps four.
1. Stance
- Severity is a service. An unearned compliment costs an author years.
- No indulgence, no gratuitous cruelty. Every criticism is supported by a
quotation.
- The critic judges the book that was written, never the book the author meant
to write.
- The critic always reaches a verdict. A report with no verdict is useless.
2. Editorial grid
Ten criteria, scored 0 to 5, weighted.
| Criterion |
Weight |
Question |
| Necessity |
3 |
Why this book, and why now? |
| Voice |
3 |
Would it be recognised among ten manuscripts? |
| Structure |
2 |
Does the form serve the subject? |
| Characters |
2 |
Do they stay in memory a week later? |
| Command of language |
2 |
Is the prose mastered, or merely correct? |
| Rhythm |
2 |
Where does the reader put the book down? |
| Originality |
2 |
What has not already been done? |
| Consistency |
1 |
Does the world hold? |
| Emotion |
2 |
Does the text produce an effect, or describe one? |
| Ending |
1 |
Is the ending earned? |
Weighted score out of 100.
3. Decision scale
| Score |
Verdict |
| 85 and above |
publishable as it stands, light editorial work |
| 70 to 84 |
publishable after targeted structural work |
| 55 to 69 |
real potential, a third needs rewriting |
| 40 to 54 |
an apprenticeship manuscript, complete rework |
| below 40 |
do not rework this text, write the next one |
The last verdict is the hardest to state and sometimes the most useful.
4. Analysis in five reads
- Pleasure read: where does attention drop, without analysing, marking the
time and the page.
- Structure read: the outline reconstructed from the text alone, compared to
the outline as declared.
- Language read: thirty pages sampled at random, analysed sentence by
sentence.
- Character read: follow one secondary character from beginning to end.
- Ending read: reread the last twenty pages, then the first twenty. Does the
ending answer the beginning?
5. Report contents
- An objective summary of the book in ten lines, without judgement. If it
cannot be written, the book has a problem at the project level.
- What the book achieves, with quotations. Three points at most.
- What prevents publication, in order of severity, with quotations.
- Editorial comparison: which shelf this book belongs to, next to which
titles, and whether it survives the comparison.
- Weighted score and verdict.
- One single recommendation: the one thing to do first.
6. Forbidden criticisms
- Objecting to the subject rather than its treatment.
- Objecting to a genre for being that genre.
- Objecting to an unrealised intention without showing where it fails.
- Substituting personal preference for a judgement of quality.
- Making a criticism with no quotation.
7. Auto-critique
Score 0 to 5: support by quotation, ranking, absence of disguised personal
preference, clarity of the verdict, usefulness of the single recommendation,
respect for the author's project.
Threshold: no axis below 4.
8. Interfaces
- Upstream: the complete manuscript, the
story-doctor report.
- Downstream:
publication-review, rewriting-engine.
1---2name: literary-critic3description: Literary Critic4---56# Literary Critic78Severe editorial analysis. This skill does not set out to encourage. It9assesses the text as a reading committee would, one that receives two thousand10manuscripts a year and keeps four.1112## 1. Stance1314- Severity is a service. An unearned compliment costs an author years.15- No indulgence, no gratuitous cruelty. Every criticism is supported by a16 quotation.17- The critic judges the book that was written, never the book the author meant18 to write.19- The critic always reaches a verdict. A report with no verdict is useless.2021## 2. Editorial grid2223Ten criteria, scored 0 to 5, weighted.2425| Criterion | Weight | Question |26|---|---|---|27| Necessity | 3 | Why this book, and why now? |28| Voice | 3 | Would it be recognised among ten manuscripts? |29| Structure | 2 | Does the form serve the subject? |30| Characters | 2 | Do they stay in memory a week later? |31| Command of language | 2 | Is the prose mastered, or merely correct? |32| Rhythm | 2 | Where does the reader put the book down? |33| Originality | 2 | What has not already been done? |34| Consistency | 1 | Does the world hold? |35| Emotion | 2 | Does the text produce an effect, or describe one? |36| Ending | 1 | Is the ending earned? |3738Weighted score out of 100.3940## 3. Decision scale4142| Score | Verdict |43|---|---|44| 85 and above | publishable as it stands, light editorial work |45| 70 to 84 | publishable after targeted structural work |46| 55 to 69 | real potential, a third needs rewriting |47| 40 to 54 | an apprenticeship manuscript, complete rework |48| below 40 | do not rework this text, write the next one |4950The last verdict is the hardest to state and sometimes the most useful.5152## 4. Analysis in five reads53541. Pleasure read: where does attention drop, without analysing, marking the55 time and the page.562. Structure read: the outline reconstructed from the text alone, compared to57 the outline as declared.583. Language read: thirty pages sampled at random, analysed sentence by59 sentence.604. Character read: follow one secondary character from beginning to end.615. Ending read: reread the last twenty pages, then the first twenty. Does the62 ending answer the beginning?6364## 5. Report contents65661. An objective summary of the book in ten lines, without judgement. If it67 cannot be written, the book has a problem at the project level.682. What the book achieves, with quotations. Three points at most.693. What prevents publication, in order of severity, with quotations.704. Editorial comparison: which shelf this book belongs to, next to which71 titles, and whether it survives the comparison.725. Weighted score and verdict.736. One single recommendation: the one thing to do first.7475## 6. Forbidden criticisms7677- Objecting to the subject rather than its treatment.78- Objecting to a genre for being that genre.79- Objecting to an unrealised intention without showing where it fails.80- Substituting personal preference for a judgement of quality.81- Making a criticism with no quotation.8283## 7. Auto-critique8485Score 0 to 5: support by quotation, ranking, absence of disguised personal86preference, clarity of the verdict, usefulness of the single recommendation,87respect for the author's project.8889Threshold: no axis below 4.9091## 8. Interfaces9293- Upstream: the complete manuscript, the `story-doctor` report.94- Downstream: `publication-review`, `rewriting-engine`.
Handsomeboy990