The written word tends to be more true than the spoken word simply because an author has more time to vet his ideas when writing. Even a sharp individual with lots of good information and the best of intentions about being truthful can't always get everything correct in casual conversational speech. There just is not enough time to think everything through from every angle. The writing process itself is a powerful tool for arriving at the truth, having a record of thought to go back over not only allows clearer communication but also acts as a type of information processing to aid in idea development. Add to this the simple fact that a written document can easily be worked over and edited for content and clarity and it is easy to see that the pen (or pencil or typewriter or word processing computer) is indeed a mighty force in society. Despite the enormous utility of written language mistakes are still made. Understanding something about the types of errors in writing and how they arise is useful in sorting out what was meant by the author and how useful the information might be.
Classes of Errors in Writing
Correctness of Information
Elimination of Errors
The most talked about type of error in writing is the typo, a simple mechanical mistake. Strictly speaking a typo is limited to something like "toxix" coming out instead of "toxic", hitting the wrong key. More broadly though typos are often considered to be any mistake in writing that is not attributed to some other cause. Misspelled words are often considered typos when the author insists that he did in fact know how to spell the word. It just did not come out right at the time of writing. Likewise grammatical errors are often passed off as typos. Again the author insists that he did know an additional punctuation mark was required; somehow the required keystroke just did not get registered.
Typos, misspellings, grammar mistakes and even sometimes using incorrect vocabulary are all structural problems with writing. These structural problems are always showing up, and as long as most of them are satisfactorily corrected they do not interfere much with the finished product. The other types of errors in writing have to do with the information itself, and how that information is presented.
If the presentation of information is done very poorly then readers will be confused to the point where the problems have to be considered errors in the writing. This can happen through excessive amounts of ambiguity, but also can be a case of important points being lost in a sea of inconsequential information. If information is simply poorly structured so as to be difficult or boring to read then the writing would simply be called useless, of poor quality, or poorly written. Boring, poor quality writing is different from writing that contains errors.
Of course an author not possessing sufficient true and correct information on the subject he is writing about is going to result in errors in his writing. This is however not the only way that incorrect or incomplete information ends up in written material.
Errors in writing can also be the result of a variety of other factors. This might seem a bit hard to understand at first. Why would a skilled writer who carefully proof reads his work end up producing written materials with factual errors when he in fact possesses the true and correct information required to eliminate these errors?
The most important class of factual errors in writing that are not the result of ignorance or lack of true and correct information are computational errors. This might sound like it applies only to mathematics, science or engineering, but the reality is that computational errors can occur in nearly all types of writing even when numbers are not involved.
Grammar problems are often a type of computational error, just not thinking the logical meaning of a sentence through carefully enough. Pieces of information that fit together in complex but systematic ways are also prone to computational errors. A small number of big mistakes in describing the way that pieces of information go together can throw the whole communication process off and prevent readers from understanding the rest of the paragraph, chapter or entire book.
The process of writing ideas down is the first step in eliminating errors. Once a set of ideas has been put to paper it can be read over and scrutinized for accuracy and impact. Reading over or rewriting something turns up computational errors, and an author is often motivated to go check some information or look for additional information after having first put ideas down in writing. Very often someone might think that they understand all aspects of a subject, and only when they try to write something down do they realize that there are holes in their understanding. This is an important part of the writing process.
In many cases it only takes some small additional pieces of information or re-evaluating the correctness of certain key pieces of information for a whole set of ideas to begin to fall into place in a more organized way that will be meaningful for readers.
When an author reads over his own work he is also often inspired to explore new lines of reasoning or new ways of looking at the subject. Following through some of these new lines of reasoning with new writing is a great way to uncover problems with the original piece of writing or to improve upon the readability and impact of the original piece of writing. Writing is a process, and can provide quite a lot of processing power for working out the meanings of complex sets of ideas.
Some errors in writing don't get eliminated during the writing process. These persistent errors are of great importance because it is these errors that readers ultimately end up having to contend with. The reasons that errors don't end up eliminated are as varied as the reasons that errors turn up in the first place.
The main reason that errors in writing persist and are not eliminated is of course continuing ignorance on the part of the author. If the author does not know enough about the subject material to identify an error then the error tends to stick around. Editorial assistance can deal with some of these ignorance based errors, but only to a certain extent. Even if someone else reads over a piece of writing and finds an error it may be difficult for the author to satisfactorily remove that error. This would of course often manifest as a difference of opinions about what is actually true, but differences of opinion about the truth are only one aspect to persistent errors. Even if it is immediately apparent to an author that an error is real when it is pointed out to him he may still find it difficult to satisfactorily remove that error. Just because a particular piece of information proves to be wrong does not mean that the author (or anyone else for that matter) is going to be able to easily come up with a way eliminate the incorrect piece of information while preserving the rest of the substantially truthful and correct piece of writing. The classic example of these particular difficulties is when an editor tries to restructure a piece of writing to eliminate an error and only ends up adding a half dozen additional errors out of his own ignorance.
Ideally an author would do whatever it takes to get any factual error out of his writing when it is pointed out to him, even if it involves re-writing the entire work. This can however get pretty expensive, particularly when the work in question is an entire book that many people have been counting on making a killing selling. In the interest of getting the book sold and getting paid many people in publishing industries have for this reason have focused more on the aesthetic appeal of writing than the actual factual accuracy. As long as it sounds good it is likely to sell well.
Then there are also persistent errors in writing that neither the author or editor know anything about. These errors are often based on widespread misconceptions, nobody notices that there is something wrong with the particular piece of writing because similar mistruths exist throughout society at large.
When someone writes that 8 or 10 or even 12 psi of boost on a 10:1 compression ratio port injected gasoline engine works great the validity of the information is not questioned because lots of people are saying the same thing. As long as the fuel suppliers deliver ridiculously supper high pressure specialty race gas that 8 psi of boost on a 10:1 port injected engine does appear to work well and few people have any clue as to the severe problems these errors are likely to cause.
In this case the error is not necessarily the fault of the author, or editor or even the readers. They have been set up for it. The fault lies with the fuel suppliers for delivering specialty high pressure race gas and calling it standard motor vehicle fuel, and also with automakers for producing engines that they well knew would not run on standard motor vehicle fuels.