SCRIPT

Hello everyone, how are you? I’m going to present a chapter from the book “A sourcebook for
basic writing teachers” by Theresa Enos, which was published in 1987. The chapter’s name is
“Constructing taxonomies for errors (or can stray dogs be mermaids?)” and it was written by
Glynda Hull.

I chose this paper because I found it very useful for my research paper in which I talk about
“The editing process in less skilled and more skilled college writers”. I will get into detail why
the editing process is related to the correction of errors or the detection of errors later, but
first I’m going to tell you a little bit about the writer of this chapter.

“Constructing taxonomies for error” was written by Glynda Hull, who is a:

-Professor of Education in Language, Literacy, and Culture at the Graduate

School of Education, University of California, Berkeley, and
-Visiting Research Professor in Steinhardt School of Culture, Education, and
Human Development at New York University.

She has published more than 100 articles, chapters and books.

She is interested in:

- The teaching of writing;

- digital technologies and their uses in schools;
- adult literacy and the changing contexts and requirements for work; and
- Community, school, and university partnerships.

Before getting into the explaining of this paper, I decided to provide the definition of
taxonomy, which is a concept that I will be repeating a lot throughout this presentation and
which was unknown for me before starting the research for my paper. The Cambridge
Dictionary defines taxonomy as “a system for naming and organizing things into groups that
share similar qualities”. So basically, it is a scientific way for saying “classification”.

To start developing the concept of taxonomy, Glynda Hull decides to introduce this topic for us
readers with an example that Jorge Luis Borges included in 1952’s “otras inquisiciones”. The
example consisted in an unknown compiler of an old Chinese encyclopedia who once divided
the world’s animals into very strange and unrealistic categories which were:

a. Animals that belong to the Emperor

b. Embalmed ones
c. Those that are trained
d. Suckling pigs
e. Mermaids
f. Fabulous ones
g. Stray dogs
h. Those that are included in this classification
i. Those that tremble as if they were mad
j. Innumerable ones
 k. Those drawn with a very fine camel’s hair brush
l. Others
m. Those that have just broken a flower vase
n. Those that resemble flies from distance

When we read this example we cannot help think that we are in the presence of a very
ridiculous classification or that we are reading some nonsense, but of course we are talking
about a Borges’s classification so it wouldn’t be nonsense and what he was doing with this
example was using this to remind us that “there is no classification of the universe that is not
arbitrary and conjectural”.

This inspired Glynda Hull for this paper and she had 2 purposes: (1) like the chinese
encyclopedist, she wanted to provide a taxonomy, a classification; AND ALSO (2) to specify
what this taxonomy cannot do. AND HER TAXONOMY OF COURSE HAS TO DO WITH WRITING,
i.e. it treats and orders a set of phenomena which exist only as a by-product, as a consequence
of written language representation and which generates attention only by its presence, not by
its absence.

Glynda is focused on “sentence-level” error in writing – commonly known as mistakes in

grammar, punctuation, spelling, usage and syntax.

- Why is she interested in sentence-level error in error? Because it has not been
customary in composition research to invent taxonomies for the analysis of
“sentence-level” error. Instead, people use an existing classification which consists
of grammar categories or usage handbooks categories, like punctuation,
agreement and so forth.

Then she proceeds to mention some of taxonomies that have been established for the errors
students make in their compositions:

1. R. L. Lyman who in 1929 in Summary of Investigations Relating to Grammar, Language,

and Composition (where he reviewed 400 research articles on writing and speaking)
2 classifications
a) To sort errors, this classification has been devised:
 Errors in punctuation
 Errors in expression
 Errors in capitalization
 Errors in abbreviation
 Errors in spelling
 Errors in grammar
 Improper margins
 Errors in writing numbers
 Errors in word-compounding
b) A classification which was devised after counting the errors in over 3000 essays
by high school students:
 Mistakes in case
 Other misuse of pronouns
  Misuse of verbs, adjectives and adverbs, prepositions and conjunctions
 Ungrammatical sentence structure
 Lack of clearness
 Mistakes in punctuation
 Misuse of the apostrophe
 Mistakes in capitalization
 Omission and repetition
 Misspelling
 Quotation mark
 Miscellaneous
2. In 1980, Lillian Bridwell’s taxonomy for sentence-level error was similar to those
mentioned in Lyman’s review. She lists:
o Spelling
o Punctuation
o Capitalization
o Verb form
o Abbreviations
o Symbols and contractions vs. full form
o Singular vs. plural
o Morphological conditioning

Of course, there’s a problem with these traditional taxonomies: Glynda argues that the act of
imposing a system of categories like the ones I just mentioned is a matter of some judgement
and choice, i.e. these categories are subjective. E.g. some categories could overlap, such as
“quotation marks” are certainly a part of “mistakes in punctuation”, or in a sentence we could
identify errors that could fulfill different categories at the same time.

By saying this, she doesn’t want to imply that taxonomies don’t matter or that they shouldn’t
be devised. It is the complete opposite; she says that “how we divide things up circumscribes
how we view things”. Taxonomies should allow us, teachers or students, to view sentence-
level errors in helpful and interesting ways.

Glynda Hull offers a new taxonomy for sentence-level error that focuses on the
phenomenology of error in rereading texts, i.e. a taxonomy which is turned upon the editing
process. She wanted to sort errors according to what happens when a writer, acting as an
editor, perceives an error in a text.

Her classification includes:

 CONSULTING ERRORS: the editor can consult her knowledge of the conventions of
written language. This category is based on the writer’s being able to call up a rule.
E.g. when an editor sees the sentence “The omniscient one had spoken, and our class
grudgingly accepted this geniuses words” in an essay, he might recall a rule for
forming the possessive in order to correct the mistake
 INTUITING ERRORS: the editor might recognize an error relying on a general sense
that something is wrong with the text, that it doesn’t seem right.
 E.g. the sentence “I think that by adding to the confusion was the fact that Dr. Lesgold
had an inferiority complex”. We hear it and it sounds wrong, so an editor might
experiment until he can make the wording sound better.
Here, instead of beginning with a verbal representation of a particular rule, an editor
would begin with a dissatisfaction that something is out of synch based on his
knowledge of the structure of the language.
 COMPREHENDING ERRORS: an editor can notice that something is wrong with the
meaning of the text, i.e. he can attend to semantics.
E.g. In the paragraph:
“Dr. Lesgold was able to devise a way to deprive our senior class
of the highlight of senior year. However, every year the biology
class had gone to an anatomy lab to see the medical students
work on cadavers. My senior year would have been no exception,
but Dr. Lesgold scheduled the trip on Senior Skip Day”
An editor might notice that the however doesn’t make sense in the context of the
paragraph. Here, the editor doesn’t need to call up a rule or experiment based on a
general sense that something is wrong; they pay attention to what the text means or
what it fails to mean.
RESEARCH
And if these categories do represent different correction strategies, one would expect
a difference in how writers respond to errors that represent these categories in
written texts. To test this expectation, she conducted a study which consisted in:

 her college students, who were divided in experts and novices, had to correct
the errors they saw in…
 essays they handed in to Glynda and also some essay samples she provided
 she categorized the errors the writers made and corrected as consulting,
intuiting and comprehending
 students corrected the errors and they were recorded when doing it.
RESULTS
 Results on the strategies:
a. In the consulting category, novices were poorest proportionately at
correcting errors and experts were much better here (23% vs. 49%). One
would expect novices to have the most trouble with this category, for it
turns on knowing written language conventions, and novices would
presumably have had the least familiarity with such conventions
b. In the comprehending category, experts were poorest proportionately at
correcting errors, but novices were somehow better here (31% vs. 35%).
One would expect experts to detect and correct mistakes that have to do
with meaning more readily than novices.
c. In the intuiting category, novices made more mistakes than experts (38%
vs. 56%)
 Results on the different types of writings/tasks: the editors were much better
at correcting both consulting and comprehending errors in someone else’s
writing than in their own (for consulting: 47% vs. 24%; for comprehending:
 52% vs. 14%). However, they could correct intuiting errors about equally well
in both tasks (48% vs. 47%)

Now, we have reached the conclusions and the taxonomy Glynda Hull has proposed could be
used in research on sentence-level error and it could reveal that there are differences in the
kinds of errors that students of various ages can correct.

She also states that the categories she has proposed are not free of problems:

- As the other categories of error, there is difficulty in deciding which elements

belong to which category
- In some cases categorizing some errors as consulting, intuiting or comprehending,
will seem arbitrary
- This taxonomy is context-dependent

However, using this taxonomy could be more insightful  the whole notion behind this system
is that there are certain strategies that we have available to us when we edit. And assigning a
particular error to a particular strategy or category also tells us something about our own
notions of editing and what actually happens at the intersection of another editor and that
error. This is important for us as teachers or future teachers as it helps us know what our
students need to learn to edit.

We can think of editing as these 3 operations: comprehending, consulting and intuiting, i.e.
sometimes editing begins with an act of comprehension as a writer attends to the meaning of
a text. Sometimes it begins with a verbal representation of a rule. And sometimes it begins
with a writer’s sense that something is amiss, although he cannot specifically name what it is.

For a long time, we’ve made the mistake of thinking about editing as a skill which is easy for
everyone – because we’ve already acquired it, because for us it is often an activity that is
internalized and automatic. We’ve thought of editing as a kind of simple pattern-matching: an
error on the page triggers the remembrance of a general rule; we see an error and we access
its correct for, card-catalog fashion. Thinking of error this way hasn’t allowed us to be very
imaginative or helpful in our attempts to teach students to edit.

So, finally, this chapter ends with Glynda’s suggestions. She proposes:

 to represent editing to students as a procedure, as an activity with several facets

 to demonstrate that there are rules that writers must know in order to correct their
texts but editing can begin with an act of comprehension or with a general
dissatisfaction that something is amiss in a text as well.
 And last but not least, to make it possible for students to practice the procedure of
editing by attempting themselves to locate and detect errors in whole texts.