When someone says that writing is good or bad, better or worse, is it merely a way to talk about whether something is popular, or interesting to you? Or is there more to it than that? Compare the two passages:

Where dips the rocky highland
Of Sleuth Wood in the lake,
There lies a leafy island
Where flapping herons wake
The drowsy water rats...
-From "The Stolen Child", by W. B. Yeats


The dog saw a frog
On a log in the bog.

I'm less interested in why Yeat's poem is technically better (unless your argument is that technical quality is the only objective measure of quality in writing). Rather, is there writing that is better or worse? Is it possible to quantify such a thing?

(Obviously, this will be seen by some as primarily opinion based - since I'm asking what exactly opinions are or should be based on.)

2 Answers 2


Writing is judged both objectively and subjectively.

Bad grammar, bad spelling, generic labels, clichés, etc can be objectively identified. Long passages of uninterrupted dialog can be objectively identified, long preambles without any action can be identified. Deus ex Machinas can be objectively identified, and explained. There is a reasonable chance a computer program can identify bad writing. A sophisticated and complex program, but using simple "if/then" programming, it doesn't have to be an artificial intelligence or quantum computer.

Much of bad writing is objective, and scientific: Don't do that, most readers don't like it, most agents and publishers don't like it, it just isn't popular.

On the other hand, most good writing is subjective. A computer program cannot tell you if the word choices in a poem combine to elicit an emotional response. They can't tell you if a newly imagined plot hook is compelling. They can't tell you if the actions you invent for your villain produce suspense or horror of sympathetic anger and grief with your MC. Computers cannot tell if a sex scene will be arousing to a reader, or painfully stupid, or laughable, or if it is too long, or if it is completely unrealistic. It takes human readers to judge whether your characters feel like real people to them, or feel like cardboard caricatures.

The same thing for jokes and humor. It takes a human. If you write scenes you hope are truly funny, there is no objective way to tell if they are. People laugh, or they don't, or they roll their eyes.

Breaking immersion is mixed; an anachronistic reference in an ancient setting can objectively break immersion, but some purple prose can also break immersion, as can an inappropriate word choice, and those can be subjective.

We can follow objective rules and logic to determine that something is bad writing.

But not to identify ALL bad writing, just because the grammar and spelling is good, and the dialogue is broken up with action or setting, and it doesn't use clichés, etc, doesn't make it GOOD writing. It can still be bad.

We cannot follow objective rules and logic to determine that something is good writing, and good writing can also overcome some of the technical gotchas that would normally classify something as bad. I've seen typos in great books, for example.

Good writing generates feelings, imagery, a sense of wonder. Those are all subjective judgments, there just is no rule for "this one line will make people want to cry," or "this line will make people laugh out loud."

  • I agree; especially with today's level of technology. I'd stipulate that it might be true that that will not hold and we're all put out of ever having a job by AI overlords in the future. Brains are meat machines, and a silicon machine should be able to catch up, eventually. Model, not algorithm most likely.
    – Kirk
    Nov 11, 2019 at 22:02
  • 1
    Tools like 'Grammarly' are beginning to implement a feature that suggests the tone of your writing. With this, it could be argued that we're approaching an era where software could indeed identify styles, moods, emotions, arousal, confusion, and the likes in different types of writing. However, I overall completely agree with your answer.
    – storbror
    Nov 12, 2019 at 9:13
  • 4
    @Kirk I am a research scientist, involved in artificial intelligence. I agree brains are meat machines, and anything the meat can do can likely be replicated with future technology. But IMO AI may be centuries from being able to write emotionally evocative original poetry, novels, or fiction. I don't think we (AI researchers) have any grasp on emotions at all, we don't know how they work in the brain, we can't even agree on what basic consciousness is, much less what constitutes sexual attraction or parental love. Writers are not in danger.
    – Amadeus
    Nov 12, 2019 at 11:03
  • @storbror Grammary processing tone is not an argument for it recognizing whether a description of a woman's first romantic kiss will make readers wince or swoon or pretend to gag.
    – Amadeus
    Nov 12, 2019 at 11:07
  • @Amadeus-Reinstate-Monica, I was simply informing that software was beginning the journey into similar territory. I'm not saying it would know how people would react to it - simply that it might attempt it 'soon'.
    – storbror
    Nov 12, 2019 at 13:10

The question is invalidated by its wording. The 'quality' of any piece of communication can only be judged against the instigator's intent.

If bad grammar is used to provoke a calculated reaction from the recipient - then bad grammar is excellent.

But to judge the quality of a a piece of writing the reader would have to know the author's intent.

Take, for example: "Green Eggs and Ham". On one level it's total garbage - nonsensical even. However, it does have a catchy, repeitive quality that children may appreciate. So the story gets a pass.

Now consider the parameters under which the story was written: "Any idiot can write a kid's story. You only need to know fifty different words."

"Green Eggs & and Ham" contains exactly fifty different words, making it a high-quality, highly-skilled piece of work by a talented individual - I doubt I could do it.

  • More importantly, Dr. Suess was able to teach a rather important lesson about dismissing outwardly unpleasant appearing experiences without first experiencing it yourself. It's rather hard for the target audience to understand this concept as any parent of a fussy eater will tell you, but Green Eggs and Ham still provides a very important lesson and it's escalation of bizarre conditions can it's lesson doesn't apply to just food. I maintain that the Lorax is a far better discourse on conservation than anything said by any politician with a green agenda.
    – hszmv
    Nov 14, 2019 at 19:22

Not the answer you're looking for? Browse other questions tagged or ask your own question.