Not long ago I stumbled across a post on a Bible software forum that taught me a new word: "pangram." I had to Google it to find out that a pangram is a sentence that includes every letter of the alphabet. A classic example is that odd sentence used to display all the characters of a typeface: "The quick brown fox jumps over the lazy dog." The forum post I read cited Deuteronomy 4:34 and Zephaniah 3:8 as examples of pangrams in the Hebrew Bible, then asked how to find a pangram in the Greek New Testament or Septuagint.
Now, I have no idea why the person who wrote the post wanted to find such sentences in Greek, but I could certainly see it being useful in an introductory Greek course. As students are trying to learn the Greek alphabet, the instructor could assign a few sentences to read which would force them to deal with every letter.
Of course, regardless of whether such a search has any practical value, it is an interesting challenge, and it got me wondering how such a search could be constructed using Accordance. Here's what I came up with:
By using the asterisk wildcard on either side of each letter of the alphabet, I told Accordance to look for any word containing each letter. By joining those together with the AND command, I told Accordance it must find a verse containing at least one word with each letter of the Greek alphabet. Note also that I've enclosed each search term in quotation marks to make sure I am searching inflected forms (the words as they appear in the text) rather than lexical forms (the dictionary form of each word).
This search finds two verses in the New Testament which contain all 24 letters of the Greek alphabet: Matthew 5:30 and Revelation 2:10. However, Revelation 2:10 does not contain a true pangram, since the entire alphabet is contained in two sentences rather than just one. To make sure we find all the places where a single sentence contains all the letters of the alphabet, we simply need to click the plus icon to the right of the search field, then set the first pop-up to Scope and the second pop-up to Sentence.
When you hit return to run the search, Revelation 2:10 is eliminated, while a number of long sentences spanning more than one verse are added.
By the way, when I ran a similar search on the Hebrew Bible, Accordance found 17 verses, including Deuteronomy 4:34. However, it did not find Zephaniah 3:8, the second example given in the original post. The reason is that Zephaniah 3:8 has all the letters except sin. Since it does contain the letter shin, whether or not you consider Zephaniah 3:8 as a pangram depends on whether you treat sin and shin as one letter or two.
So there you have it. If you ever need to find sentences with every letter of the alphabet (and who doesn't?), Accordance can do it easily and accurately.
As a writer, I have often known the mockery of the blank page. It glares at you, laughing at your struggles to fill it with something meaningful.
As Accordance users, we are constantly presented with a blank search box, ever ready to be put to use. Experienced users know exactly what to do with it, but new users may experience something of the mockery I feel when trying to fill a blank page. They may conceive of a search they would like to do, but how to construct it? They know Accordance is capable of much more than simple word and phrase searches, but how do they go about learning all the Boolean commands, wildcard symbols, and other tools that make such power-searching possible?
Thankfully, everything you need to fill in that blank search box—and I mean everything—is always readily available through one of the menus at the top of the screen. Can you guess which one?
As I'm sure you guessed, the aptly named Search menu presents you with everything you need to fill in the blank. Not sure what word to search for? Choose Enter Words…. Want to search a Bible with Key numbers for a particular key number? Choose Enter Key Numbers…. If your search text is a grammatically tagged Greek or Hebrew text, these menu items will appear as Enter Lexical Forms… and Enter Inflected Forms….
Beyond simply helping you enter words and key numbers, the Search menu also includes submenus listing every search command (AND, OR, NOT, etc.), every wildcard symbol, and (in the case of tagged texts) every grammatical and syntactical tag. You don't have to memorize these options or go digging through documentation even to realize they're available; just go to the Search menu and browse through the submenus. Not sure what a command or symbol does, but want to try it out? Simply select it from the menu to insert it into the search box.
By making all these options readily available, Accordance does its best to eliminate the potential mockery of the blank search box. And while there are still aspects of these commands and symbols which need to be learned, you always have them listed in a convenient place whenever you need them.
If you've never paid much attention to the Search menu, you now understand it's importance. Just remember to look there whenever you need help filling in the blank.
In yesterday's post, I gave Hebrew students a technique for garnering extra credit by searching for geminate verbs: verbs like הלל in which the second and third letters of the lexical form happen to be the same. To do that, we used the question mark wildcard to represent each letter in the lexical form, and then we specified that the third letter had to be the same as the second by placing (=2) after the third question mark. This ability to specify that different letters must match was added specifically for these kinds of original language searches, but we can also use it to have a little fun with our English Bibles.
"Palindrome" is a fancy term for a word that reads the same way both forward and backward, such as the name "Hannah." (There's a popular children's book with the title Hannah is a Palindrome.) Since we can have Accordance find words with matching letters, we can easily find all the palindromes in the Bible.
Let's start with three-letter palindromes. To find those, enter three question marks, and place (=1) after the third question mark. Like this: ???(=1)
To find a four-letter palindrome, you would enter four question marks, placing (=2) after the third question mark, and (=1) after the fourth: ???(=2)?(=1)
For a five letter palindrome, simply add another question mark to the beginning of the previous search: ????(=2)?(=1). The third question mark does not need to match any other letters.
You get the idea. If you want to get a list of all the palindromes in the Bible, try this search: ???(=1) <OR> ???(=2)?(=1) <OR> ????(=2)?(=1) <OR> ????(=3)?(=2)?(=1). I tried searching for seven-, eight-, nine-, and ten-letter palindromes, but didn't find any in the translation I searched. Can anyone else find a Biblical palindrome longer than six letters?
Once you've done this search, select Analysis from the Stats & Graphs icon to get a list of all the palindromes in the Bible.
Note that because the question mark can represent any letter or number, our search found numbers as well as words. Can anyone come up with an easy way to exclude the numbers?
Update: Not ten minutes after I posted this, my youngest daughter pointed out that today's date is 3/1/13 (when written using the American convention of month/day/year). It would appear this post on palindromes was more timely than I realized!
In yesterday's post, I offered you Greek students a method for garnering extra credit from your professors: Approach them on Friday and ask for extra credit if you spend the weekend finding all the hapax legomena in the Greek New Testament. Then do a simple search in Accordance, open an Analysis tab, and print! (Be sure to look tired and disheveled when you turn it in on Monday.)
Now, you Hebrew students could use extra credit too, so far be it from me to leave you out. Here's a search that is sure to impress your first semester Hebrew prof. In Hebrew, there's a class of verbs known as geminates. These are verbs in which the second and third letters of the lexical form are the same. Perhaps the most famous example would be הלל, "to praise." Why not ask your professor if you can spend the weekend putting together a list of every geminate verb in the Hebrew Bible? It sounds appropriately difficult—like the kind of thing only the most ambitious Hebrew student would attempt.
Now, to define this search, start by entering three question marks in the search entry box. The question mark, like the asterisk, is a wildcard symbol which can stand for any letter or number. Yet where the asterisk can represent any number of characters, the question mark can only represent a single character. For example, if I do an English search for love*, Accordance will find any word that begins with "love," no matter how many other letters it has: words like "loves," "loved," "lover," "lovers," "lovely," etc. If, on the other hand, I search for love?, Accordance will find only words that have a single letter after "love", such as "loves" and "lover." See the difference?
Because each question mark stands for a single character, whatever that character happens to be, entering three question marks into the search field means you want to find only three-letter words.
Now we need to specify that the third letter must be the same as the second letter. To do that, place parentheses after the third question mark. (Be sure to enter the right parenthesis followed by the left, since everything is reversed in Hebrew). Inside the parentheses, place an equals sign followed by the number 2. Your search argument should now look like this: (2=)???.
Placing parentheses after a question mark lets you specify which characters that question mark can be. For example, b?(ae)t will find "bat" and "bet," but not "bit" or "but," since we have specified that the question mark can only be an "a" or an "e." Returning to our Hebrew search, placing (2=) after the third question mark means that it can be any character, so long as it is the same character as that found by the second question mark. In other words, the second and third letters can be any letter, but they must be the same.
Now all we need to do is specify that the word we're searching for must be a verb. To do that, make sure the cursor is blinking to the right of the search argument, then choose Verb… from the Enter Command submenu of the Search menu. A dialog will appear enabling you to select specific grammatical details, but since we want to find any verbs, just click OK to dismiss the dialog. Accordance will then add the Verb tag to your search argument, and all you need to do is hit Return to perform the search.
As sophisticated as this search is, your results should be instantaneous. Now just choose Analysis from the Stats and Graphs icon to get an alphabetical list of all the geminate verbs in the Hebrew Bible, and print it out!
Just be warned: while this search may help you garner some extra credit, it may also raise your prof's expectations for you. You might find that he now treats you as a star pupil. And we all know what that means: you can probably count on getting called on more often in class!
In yesterday's post, I showed how to exclude certain words from a wildcard search. Yet since I did so in the context of a complex search we had built over the course of several other blog posts, I thought it would be helpful in this post to back up and explain this feature more simply and systematically.
Let's start with a simple search for the word "love." In the HCSB (my preferred translation), this word occurs 569 times. Yet this simple search does not find other forms of "love," such as "loves," "loved," "loving," etc. If I want to broaden my search to include those other forms, I can replace the "e" with an asterisk wildcard. Searching for "lov*" will find any word beginning with "lov", because the asterisk symbol represents any combination of other characters. When I perform this search in the HCSB, I get 798 occurrences.
A convenient way to see all the different words that were found is to choose Analysis from the Stats and Graphs pop-up menu.
This will open a window listing every English word found by my search. If you're searching an English Bible tagged with Strong's numbers, such as the HCSBS, the Analysis will even list all the Greek and Hebrew words behind the English. In this screenshot, I've turned off the Key number display so we can see all the English words at a glance.
Now, looking at this list, I may find that my wildcard search was a little too broad, returning results I didn't intend. For example, I want to find all forms of "love," but I don't want to find things like "lovemaking," "lovely," etc. So how can I refine this search so that it is broader than my first search for "love" but narrower than my second search for "lov*"?
As I showed you yesterday, I can place a minus sign after a wildcard search to exclude the forms I don't wish to find. Thus, "lov*-lovemaking" will find every word beginning with "lov" except "lovemaking." If I want to exclude other words, I can simply add those after additional minus signs.
I can even use wildcard symbols in the words I'm excluding. For example, to eliminate words like "lovely" and "loveliness," I can add "-lovel*". My entire search argument would then look like this: "lov*-lovemaking-lovel*". And my resulting word list is as follows:
If you've been following along and trying to do this yourself, how would you now eliminate words like "lover," "lovers," and "lovesick"?
In my previous post, I used Accordance to show that an internet meme claiming that the Bible has 365 occurrences of "Do not be afraid" is clearly mistaken. First I searched my default Bible using a series of OR commands to account for possible variations in the phrase. Then I used the Search All window to search all my English Bibles. At the end of that post, I pointed out something surprising: several popular English translations (such as the KJV, NRSV, and NASB) did not show up in the search results. I even offered a ten dollar Accordance credit to the first person who could explain why that was. In this post, I want to explain why the search failed for those translations and offer a workaround.
When you search an Accordance resource, the first thing Accordance does is check every word you entered against the word list for that resource. If one of the words is not found in the word list, Accordance will actually present you with the word list so you can pick another word. If you search the NRSV for the search I created in my last post (do not be afraid <OR> don't be afraid <OR> do not fear <OR> don't fear <OR> be not afraid <OR> fear not), you'll see immediately that the NRSV does not include the contraction "don't." Thus, the search as I've constructed it is invalid for the NRSV.
When you perform this search in a Search tab, you get instant feedback that a word you entered is not found, but in the Search All window, it would be irritating to be presented with a word list for each module that did not contain a word in your search argument. The Search All window therefore only returns results for the modules that contain the search terms entered. Thus, any Bible that does not have the word "don't" does not show up in our search results, even though it may contain the other phrases entered.
If you want to make sure this search returns results for those Bibles which do not have "don't" in their word lists, you have to resort to a little trickery. By replacing the "o" in "don't" with an asterisk wildcard, Accordance will accept the search, since these Bibles have plenty of words which begin with "d" and end with "nt" (Accordance ignores the apostrophe). Since words like "descendant" and "different" are not likely to be coupled with "fear" in a single phrase, this broader search returns few if any false hits. It simply satisfies Accordance's requirement that a search term match at least one word in a module's word list. With that requirement satisfied, the Search All will now return results for those Bibles which do not contain the word "don't", including the NRSV, KJV, and NASB.
In my next post, we'll examine this wildcard search more closely and see what, if any, false hits might have been found. Then I'll show you another trick to eliminate even those.