Negating Grammatical Characteristics
All last week, we saw how Accordance enables you to search grammatically-tagged original language texts without having to learn an arcane set of codes or commands. First, we saw how Accordance uses an easily understandable and transparent search syntax, so that you can just type the grammatical characteristics you want to find. Next we looked at Accordance's simplified dialog box interface which lets you choose the grammatical details you're looking for. Rather than a confusing array of checkboxes and options, Accordance presents you with a clean and uncluttered series of pop-up menus. Finally, we examined that dialog box more closely, showing that by bending the rules a bit, we were able to create an interface which makes simple selections simple, and which can easily be extended to accommodate far more sophisticated searches.
Today, I want to talk about something which is incredibly simple in Accordance, but which apparently is very difficult to do in other programs: namely, specifying the forms you don't want to find. For example, let's say I want to find all Greek nouns appearing in the "oblique cases."
Since the oblique cases are all those other than the nominative, the simplest and most logical way to construct such a search would be to indicate that you want to find every noun which is not in the nominative case. To do this in Accordance, go to the Search menu and choose Noun from the Enter Grammatical Tag submenu. In the dialog box, choose Nominative from the Case pop-up menu. You'll notice that when you do, a checkbox appears to the left of the Case pop-up menu. Underneath the column where this checkbox appears you'll see a label which reads NOT. Any checkbox which appears in this column is the NOT checkbox for the category it appears beside. To negate the selection in the pop-up menu, just check the box.
When you click OK to dismiss the dialog box, the following search syntax will be entered into your search window: [NOUN -nominative]. Note that once again the search syntax is easily understandable. The only code used here is the minus sign, which is a pretty obvious indicator of negation. When you perform this search, Accordance will find all nouns in any case other than the nominative. If you wanted to extend this search to look for any part of speech in the oblique cases, you could simply replace NOUN with ANY and rerun the search.
Spoiled as I am by Accordance's simplicity, I was recently surprised to discover that other programs make this much harder than it should be. In one high-end program, I could find no easy way to negate an item, while in another, the user must enter a GREP-style regular expression in conjunction with arcane parsing codes. In this way, something which should be trivial becomes a significant challenge.
Throughout this series of posts, I've been explaining the very basics of the Accordance interface for searching grammatically-tagged original language texts. These are all interface conventions we established way back in 1994 when we released Accordance 1.0. Fifteen years later, users of other programs are still having to do things the hard way. Meanwhile, Accordance users are happily studying the biblical languages without having to learn a computer language.