Q. What does it mean that Word 98 has a "natural language" grammar
A. The grammar checker in Word 98 can do a more comprehensive and
accurate analysis (also known as "parsing") of the submitted text,
instead of simply using a series of heuristics (or pattern matching)
to flag errors. The Word 98 grammar checker does text analysis at a
syntactical level and at a deeper, logical, level to understand the
relationship between the actions and the people, or things, doing
those actions. For example, the Word grammar checker analyzes the
following complex sentence
The legend says that that Kingdom was created by three ancient
magicians, whose magical powers governed the world and made them
immortal and all-powerful.
and rewrites it from the passive to the active voice for clarity,
while setting off the relative clause between commas:
The legend says that three ancient magicians, whose magical powers
governed the world and made them immortal and all-powerful,
created that Kingdom.
Q. Who developed the Word 98 grammar checker?
A. The grammar checker is fully developed and owned by Microsoft.
Q. What are the key differences between Word 98 grammar checker and
other grammar checkers?
A. One of the cornerstone differences between the grammar checker in
Word 98 and other grammar checkers stems from the fact that the
grammar checker in Word 98 uses advanced parsing techniques to
understand the sentence structure while the other grammar checkers
rely mainly on "pattern matching." By pattern matching, we mean that
the program uses a technique that matches the checked text against
patterns of text stored in an internal database. Following are some
sentences that highlight the superiority of the natural language
grammar checker in Word:
- He never learned to swim, or did he want to.
The Word grammar checker corrects "or" with the appropriate
- She encourages Stephen more than Elisabeth.
The Word grammar checker corrects the sentences by proposing two
possible new sentences to make the original meaning less
- They wanted for us to move to Alaska.
Word grammar checker corrects the sentence by removing the
Q. What are the files names of the grammar checker files and where
are they installed?
A. Word (or Office) Setup installs the grammar checker by default. The
English grammar checker is comprised of two files:
MS Grammar, and MS English Grammar Dictionary
both installed in the Microsoft Office 98:Shared Applications:
Proofing Tools folder.
Q. How much memory do I need to have on my computer in order to run
the grammar checker automatically?
A. Word will enable the grammar checker automatically if your computer
has sufficient available memory. The method of grammar checking that
is enabled when you set up and first start Word depends on
the amount of available memory on your computer.
Manually Check Grammar (8 MB or More with Virtual Memory turned on):
To run the grammar checker when you click Spelling And Grammar on
the Tools menu, your computer must have more than 8 megabytes (MB)
If you have less than 8 MB, the grammar checker is turned off by
default when you first start Word.
Automatically Check Grammar (12 MB or More):
To run the grammar checker constantly (to display grammatical errors
with wavy underlines), your computer must have at least 12 MB of
physical RAM. If your computer has less than 12 MB of RAM, the Hide
Grammatical Errors check box is selected when you first start Word.
To turn on the automatic grammar checking, click this check box
to clear it (on the Tools menu, click Spelling and Grammar, and
click the Options button).
Note also that for all Western languages other than English, the
automatic grammar checker is turned off by default. (The English
grammar checker is shipped with all versions of Word.)
Q. What are the registry entries for the grammar checker?
A. Grammar is registered in the MS Proofing Tools (PPC) section of the
registry. Word 98 always looks in the PPC section of the registry
first to avoid picking up a 68K grammar checker. English PPC
grammar checker has a new file type (MSGR) so that Microsoft Word
for the Macintosh version 6.0 (if it's also installed on the
machine) cannot load it.
Q. Why does the grammar checker flag words that shouldn't be flagged,
and why does it provide suggestions that are incorrect?
A. In general, the grammar checker will incorrectly mark words or
propose incorrect suggestions, when the parser (that is, the grammar
checker component that analyzes the linguistic structure of a
sentence) cannot determine the correct structure of the analyzed
Although state-of-the-art in its category, the grammar checker
(just like any other commercially-available grammar checker program)
is not perfect. Therefore, when you use the grammar checker, you can
expect some amount of "false" or "suspect" flagging and subsequent
Q. Why can't the grammar checker spot mistakes in the phrase "We went
two too stores, to . . ."?
A. The grammar checker is designed to catch the kinds of errors that
ordinary users make every day. You will always be able to make up
sentences that will confuse the grammar checker.
Q. When the grammar checker is running in the background (wavy
underlines), why does it flag errors in a different order than when
I click Spelling And Grammar on the Tools menu (run the grammar
checker in the foreground)?
A. With the background grammar checker, trying to achieve a logical
left-to-right flow is not as critical as it is for the grammar
checker that you run manually (click Spelling And Grammar on the
Tools menu). Therefore, for the background grammar checker, the
error marked first is always the one that returns a suggestion,
regardless of its position in the sentence.
Q. Why is Ignore All not working as expected? For example, if I click
Ignore All for this sentence labeled as fragment
After serving lunch.
in the same grammar checker session, the grammar checker stops on
other sentences that are labeled as fragments, for example:
Over my dead body.
The grammar checker categorizes (internally) these two sentences as
different types of fragments, and in the above examples, grammar
checker is ignoring one of those types, but not the other. Thus the
inconsistency of how Ignore All works.
Q. Why aren't mistakes flagged in left-to-right sequence? For example:
But the army, however, went on with it's plan.
A. In most cases, the Grammar checker tries to flag errors from left-
to-right. However, In some cases this is not possible because the
grammar checker wants you to correct the most logical mistake first
(this mistake may not be the first mistake). In this case,
punctuation or spacing mistakes are flagged before specific or
confined grammar checker mistakes.
Q. Why are some passive sentences flagged and rewritten by grammar
checker, while others are skipped?
NOTE: This problem occurs with other rules in addition to the
For example, the following passive sentence is not flagged:
The term of this Agreement shall commence on the Effective Date
and shall continue until terminated by Volcano Coffee in writing
at any time, with or without cause.
A. For certain types of sentences, when there is no clear syntactic
subject, the grammar checker will not attempt to flag the sentence.
Q. When I hold down CONTROL and click a grammar error (an error
marked with a wavy underline), why doesn't the shortcut menu display
the same options that are available in the Spelling And Grammar
dialog box? For example, if an item is flagged, but if the grammar
checker does not provide a suggestion, the only options available
are to ignore the sentence (and possibly miss other errors in that
sentence) or to click the Grammar command to invoke the Spelling And
Grammar dialog box.
A. For the background mode (wavy underlines), the grammar checker uses
a simplified interface. If you want to view all the possible errors
in a given sentence, you must click Grammar on the on the shortcut
Q. Why do some pairs of words that are commonly confused work in one
direction only? For instance, in the grammar checker both flea and
flee are flagged as commonly confused words, but with the pair your
and you're, only the word "your" is flagged as a commonly confused
A. The grammar checker handles some commonly confused word pairs in a
unidirectional way to simplify the problem for the parser. The
grammar checker is designed this way to reduce the number items that
are flagged by the grammar checker but that are not true grammatical
Q. Why is it that when a sentence is flagged as being too long, that's
the only advice given for the sentence?
A. Long sentences are often difficult to read both for people and for
the grammar checker. The grammar checker is not sophisticated enough
to detect grammatical errors in long sentences. If you are in doubt
about the grammatical accuracy of a long sentence, you should break
it up into smaller sentences.
Q. Why does the grammar checker ignore text enclosed in quotation
marks? For example, this text is ignored
He said, "what on earth were you thinking of?"
while in the following text, "what on earth" is flagged:
He said, what on earth were you thinking of?
A. The grammar checker assumes that text in a direct quotation should
not be critiqued.
Q. Why does the grammar checker ignore text in subdocuments such as,
headers, footers, and annotations?
A. By design, the grammar checker does not analyze text in headers,
footers, or annotations. Headers and footers typically do not
contain complete sentences. Similarly, annotations may be written in
sentence fragments, and are not suitable for grammar checking.
Q. Why can't I set options such as the length of sentence?
A. This option is built-in to the selected writing style. To change the
acceptable sentence length, change the selected writing style. The
grammar options that are built in to the writing style include:
- Length of long sentence
- Allowable number of noun modifiers
- Allowable number of consecutive prepositional phrases
- Allowable number of words to split infinitive
- Use of the first person pronoun
NOTE: Available in the grammar checker only when the Technical
style is selected.
The following table lists the specific values for the invisible
Invisible Option Casual Standard Formal Technical Custom
Length of long
sentence 50 50 40 40 40
Allowable number of
noun modifiers 5 4 3 3 4
Allowable number of
phrases 5 4 3 3 4
Allowable number of words
to split infinitive 3 2 1 1 2
critiques off off off on off
Q. What do the grammar statistics mean?
A. The Flesch Reading Ease calculates how easy it is to read the
document. The higher the score (on a scale of 0 to 100), the easier
it is to understand the document.
The Flesch-Kincaid Grade Level provides the writer of the document
with a value that indicates the minimum education level required for
the reader to be able to understand the document. The lower the
score, the easier it is to understand the document (scale is 0 to
What formulas are these statistics based on?
The Flesch Reading Ease score is based on the number of words in
each sentence that is grammar checked, and the average syllable per
word. The Flesch Reading Ease score rates text on a 100-point scale;
the higher the score, the easier it is to understand the document.
The formula for the Flesch Reading Ease score is
206.835 - (1.015 x <ASL>) - (84.6 x <ASW>)
where <ASL> is the average sentence length (number of words/number
of sentences) and <ASW> average number of syllables per word (number
of syllables/number of words).
The Flesch-Kincaid Grade Level score is very similar to the above,
and it measures readability as a grade level.
The formula for the Flesch-Kincaid Grade Level is:
Grade Level = (.39 x <ASL>) + (11.8 x <ASW>) - 15.59
Who uses them?
Various government agencies require that the readability of specific
documents or forms meet specific readability standards. For
example, some states require insurance forms to have a specified
Q. How many words and phrases are in the grammar dictionary?
A. The grammar dictionary includes approximately 99,000 words and
phrases, in their uninflected form (that is, this number does not
include words such as "went," "children," and so on, which are the
inflected forms of "go" and "child."
Q. What is the grammar dictionary based on?
A. It is based on the "Longman Dictionary of Contemporary English," and
the "American Heritage Dictionary of the English Language," third
Q. How is the English grammar checker different if I run it on U.K.
English text versus U.S. English?
A. The difference between proofing UK English text and US English is
primarily in the spelling variances of the words in the two
languages, for example, "colour" as opposed to "color." These
variances do not have any effect on grammar.
The vast majority of the grammar rules apply to all English text
(U.S. and U.K.). However, there are just a few grammar rules that
differ depending on the selected language:
a. Plural premodifiers that are very commonly used in U.K. English,
are not flagged for U.K. English, but are for U.S. English as in
the following example:
This is a trades union. We have done all the
b. Subject-verb agreement with collective nouns where the verb is
used in the plural form are not flagged in U.K. English, but
are flagged in US English. as in the following example:
The team are planning to mobilize soon.
Q. Some of the explanations don't seem to be related to the flagged
mistake. For example, in the sentence
Lets go home now.
the explanation in the grammar checker does not mention specifically
the confusable pair lets/let's.
A. The grammar explanations are intended to cover the most general
cases within each rule in order to avoid crowding the screen text.