Home

Novels

Nonfiction

Arthurian

Writing

Bio

Webrings

(As you read this, please keep in mind it first appeared in paper format! -- kdh) 

Data-Based Characters

Can Provide Relief for the Terminally Forgetful

by Kim D. Headlee
Copyright 1993, Kim D. Headlee

An editor runs an article in your favorite writers' rag gleefully detailing her "Top Ten Slips of the Pen" from the hundreds of manuscripts to cross her desk. If you've seen it once, you've seen it a dozen times.

Invariably at least half the list betrays examples of an author's forgetfulness. For instance, the blonde bombshell who in Chapter 12 suddenly (and without a trip to her hairdresser) transforms into a ravishing redhead. The U.S. Marine colonel who, halfway into the book, finds himself in Navy blue commanding the U.S.S. Enterprise. The heroine's Harvard Law degree, later mentioned as having been earned at Yale. Hilarious to read about, but devastating to the author. No one enjoys the comparison of their memory to a brick of Swiss cheese.

You eagerly devour these boo-boos, chuckling with the editor who has shared them, smug in the certainty that mistakes like this will never escape your keyboard. Didn't anyone ever warn you never to say never?

One day the unthinkable comes home to haunt you. The fifteenth rejection slip for your romance novel is a polite, personal note with a postscript:

"By the way, I liked your hero better when he had steel-blue eyes, rather than sea-green ..."

At first you can only stare at the letter in mortified astonishment. Fourteen editors had not breathed a word about this! And probably each one -- including the fifteenth -- is busily inserting your blunder into his or her next version of "How Not To Win Contracts And Influence Editors."

In disbelief you ask yourself how, in a hundred trips across the manuscript, both on-screen and off, the mistake could have craftily eluded you every time. Denial builds as you vainly attempt to blame faulty word processing software, a glitch on your hard disk, or some weird computer virus.

Sorry. No electronic scapegoats pastured here.

You locate the errant passage in Chapter 34, precisely where you put it. Tendrils of despair curl around your heart. How, you wail, could you have been so blind? What other errors are lurking between the lines? Can you ever show your face in your writers' group again?

Stash the sackcloth and ashes, my friend. All is not lost. You just need to record your characters' vital statistics (vitae, for short) in a data base.

If you don't own a computer, I respectfully ask you to stay tuned awhile longer. For what is a data base, after all, but a well-organized collection of information? Your telephone directory is a data base. So are your cookbooks and weekly television listing.

My proposed solution can be implemented with index cards. The computerized data base offers the advantages of speedy retrieval, completeness and consistency. But the manual method can be just as helpful.

Using the example of a mailing list, here is a brief introduction to basic computer data base terminology and concepts. With a firm grasp of these terms you should be able to conquer any PC data base software package on the market.

A unit of information is called a field. In a mailing list data base, name, street address, city, state abbreviation, and zip code are separate fields. A field can be variable-length, as with name, address and city, or fixed-length (such as state abbreviation or zip). Whatever the length, a field can contain all alphabetic characters (A-Z, a-z only; as in a name), integers (digits 0-9 only, such as a zip code), or alphanumeric (A-Z, a-z, 0-9, and punctuation, such as street address). Other field designations include dates (say, a birth date) and floating point (good for monetary amounts, like your book advance).

The complete set of mailing information for one person is a record. This is analogous to a Rolodex (tm) card. In this manual filing system, the first letter of the last name serves as the record's key. In a computerized data base, entries tagged with the same key (referred to as duplicates) are permissible, but not advantageous. The computer's power lies in its ability to swiftly distinguish between records to retrieve the specific one you've requested (for example, Lynn Smith rather than Lynda Smith). Data base software will permit you to define your keys (the words, names, etc. you ask the computer to "find") to be as long as necessary -- even spanning multiple fields -- to achieve uniqueness.

In this example, the mailing list itself is known as a table. A data base can contain one or more tables. These tables may have identical record structures, as in a mailing list that contains your personal correspondents and another that lists your professional contacts. Or tables can be vastly different, such as a list of telephone area codes or two-character state abbreviations. These special types of tables are helpful in preventing the ever-present, ever-unpopular data entry errors.

Now you're ready to begin building your data base of character vitae, right? Wrong.

Even the most seasoned construction foreman needs a blueprint. You must first exercise your gray cells regarding not only the information you wish to put into your data base, but how it's going to be used. Therein lies the trick to designing an effective data base.

Don't start reaching for that sackcloth just yet; this type of analysis is not as difficult as you may believe.

Certain data about your characters are going to be the same whether the book is a contemporary mainstream novel or a time-travel fantasy: name, nickname, date and place of birth, sex, height, weight, hair and eye color, occupation, and outstanding personality traits, to name a few. I also recommend including a generic alphanumeric field of at least 50 characters for recording brief comments about each character. For me, this serves as a wonderful memory-jogger for important events yet to be written.

If your project is a family saga spanning several generations and volumes, you may want to include parents' and spouse's names, date and place of death, and the volume(s) in which the character appears.

Mine is a multi-volume work of ancient historical fiction. The more than 50 characters hail from a variety of cultural backgrounds across Europe. Obviously, the modern convention for address would be meaningless. Instead I use country, province, village or town and, whenever possible, clan affiliation. A geographically grouped character list is invaluable for helping me keep everyone straight.

My task is made even more formidable by the fact that my characters' names are either Latin, Celtic, Anglo-Saxon, or Arabic. Very few are simple to pronounce, let alone spell. Were it not for the artificial sanity offered by the data base I might have dumped my project long ago. (Okay, I exaggerate. But I probably would have sprouted a lot more hoary hairs by now!)

One side benefit is the convenient means for storing "dummy" characters. Forever questing for new and interesting names, I often add promising candidates to my data base without bothering to fill in any other fields of information. Data about existing characters can be expanded and updated quite easily, without enduring the torture of rewriting index cards, or smelling like a you've just doused yourself with "Eau de Fluide Correction."

A quirk with my vitae data base is that I cannot use the date field. During the time in which my books take place dates were expressed in relation to religious festivals (Samhain Eve), political terms of office ("the fifth year of the reign of Emperor Arturus Aurelius Vetarus"), manmade events ("a fortnight after the Siege of Caer Alclyd"), and natural events (like forest fires, floods, or eclipses).

Writers of science fiction and fantasy may need fields to indicate a character's species, planet of origin, moral alignment (good, evil, neutral), skill level or rank, preferred weapon, etc. If animals are closely associated with any of the characters, as with Anne McCaffrey's Dragonriders, a field for the animal's name could prove immensely helpful -- especially if several animal names are agonizingly similar.

And the list goes on. The possibilities are corralled only by your imagination.

The point is you need to establish a scheme that works for you. Once you have sweated through the design and initial data entry, and have disciplined yourself to update the information on a regular basis, you will not be disappointed. Have you been stung by an editor's offhand remark about all your characters seeming to be tall, blonde and blue-eyed? Just query the data base. Often in no more time than it takes to flip the lid off the box of note cards, a list of names matching your search criteria will materialize on the screen.

At the very least, each search will yield a more intimate understanding of your characters -- which cannot be a bad happenstance in anyone's estimation.

Will this fancy setup require a large cash outlay? Perhaps, depending on the data base software you select.

What about a significant investment of time -- precious time that could be spent banging out yet another draft? Yes, certainly, both to design and to enter the data into your system -- not to mention time spent on the upside of the learning curve simply to become accustomed to the software.

But just imagine how much more efficient your writing process could be if your characters' vitae resided at the tips of your fingers rather than scattered at random across that vast organic hard drive called your brain. No more confusing Henry Johnson with his cousin Hank. No more embarrassing changes of eye color mid-chapter. No more snide editorial comments about your manuscript having "too many" tall, blue-eyed blondes. Experience the freedom to compose without the nagging (and time-eating) compulsion to tear through earlier writings and mounds of research notes to ensure correctness of detail.

Then, the next time your favorite writers' rag runs another "Top Ten Writers' Bloopers," you can truly relax in the assurance that none will be yours.

Publication history:

Writers Connection, May 1993