On the passage printed with the lab, 936 words of Pride and Prejudice, 149 words match a form of an irregular verb: 15.9%, about one word in six, counted in your browser. That number is true and it is also a trap. Three verbs, be, have and do, are 91 of the 149. Take them out and the share falls to 6.2%. Over the whole novel, counted once and quoted here, the same move takes 12.39% down to 5.01%. The lab also counts how much of the irregular forms the twenty commonest verbs cover, 95.3% with the three and 93.1% without, and the lesson sets those beside the two figures quoted from the novel. A list that looks concentrated is concentrated mostly in three places, and a short page makes it look more concentrated still.
Core ideas
Three ideas, and the second is the one to keep.
Every time an irregular verb appears in the text, in any form, that is one count. Had, have and having all count for have. The share is the number of those counts divided by the number of words in the text: on the passage, 149 divided by 936. It tells you how often you will meet one as you read, which is not the same as how many there are.
On the passage, be, have and do are 91 of the 149 forms, which is 61.1%. Over the whole novel the same three are 59.5% of every irregular form. So when you hear that irregular verbs fill a large part of English, most of that part is three verbs that are also the helping words of the language.
“The top twenty cover 87.66%” is true of the novel. But three of those twenty are be, have and do. With them left out, the top twenty of the rest cover 73.90%. Both figures are quoted: the page cannot recount them, because it would need the whole book. The page does count the same thing on its own passage, and gets 95.3% and 93.1%. The first quoted figure makes the list look more concentrated than the verbs you must learn really are, and the passage figures show the same effect for a different reason: a short text uses few verbs.
The material
The lab’s own count on the printed passage, and beside it the count made once over the whole novel.
How much of English is irregular? There is a quick answer, and it is the wrong place to stop. This lesson gives the quick answer, counted in front of you, then takes it apart.
The lab prints 936 words of Pride and Prejudice and checks each one against the list of irregular verbs from the first lesson. For each verb it looks for every form: the base, the past, the form after have, the form after he or she, and the -ing form. The last two are made by the spelling rules from Tense Tables, so go gives goes and write gives writing. Be and have are the two the rules cannot make, so the forms the rules miss are written out: am, is, are and being for be, and has for have. Every match counts as one irregular form. It finds 149, which is 15.9% of the words. About one word in six.
That is a large share. It is the sort of number that gets quoted when someone wants to say that irregular verbs matter, and they do. But look at what it is made of. The table under the tiles sorts the verbs by how often they appear, and the top three rows are be, have and do.
Be, have and do are 91 of the 149 forms, which is 61.1%: about three in five of every irregular form on the page. Be alone is 49 of them, and twenty of those are was. Take the three out with the menu. The other irregular verbs come to 58 forms, which is 6.2% of the words. About one word in sixteen. That is still a fair share, but it is well under half the first figure. There is a reason these three stand apart. They are not only main verbs. They are also the helping words that build the tenses from Tense Tables, so they appear whether or not the sentence is about being, having or doing anything.
The lab matches spellings, not meanings, and there are words where that matters. Left would count for leave even if it meant the side, and saw for see even if it meant the tool. In this passage the one left and both of the saws are verbs. One match is not a verb at all: by no means is counted as a form of mean, because means is the he or she form the -s rule builds. So of the 149 matches, 148 are irregular verb forms, and the share of words is 15.8% rather than 15.9%; the shape of every figure below is untouched. The table lists every verb with its count, so you can find the row for mean and check it against the passage by hand.
The same count was made once over the whole novel, with the program this page runs, over the text of the novel only: the picture captions that the 1894 printing added were taken out first. That gave 15,147 irregular forms in 122,294 words, which is 12.39%, about one word in eight. Be appears 5,858 times, 4.79% of all the words by itself. Have appears 2,339 times and do 820. Together they are 9,017 forms, 59.5% of every irregular form in the book, and the rest come to 6,130 forms, 5.01% of the words. None of those figures is recomputed on this page. The novel is about 120,000 words, some 130 times the passage, far too much to print here, so they are quoted, and they are marked as quoted wherever they appear in this lesson. What you can check is that the passage, counted in front of you, has the same shape: more than half of the irregular forms are three verbs, and the share of words falls by more than half when they leave.
Now count the other way, over the novel. Rank the irregular verbs by how often they appear and add up the top twenty. With be, have and do in the list, the twenty cover 87.66% of all irregular forms. It sounds like proof that a short list does almost all the work.
It is the number to distrust. Three of the twenty places belong to the three verbs that carry 59.5% on their own. That leaves seventeen places to share what remains. Leave the three out and rank the rest again. The top twenty of those cover 73.90%. Both figures are quoted from the one count of the novel; the lab’s table shows the same ranking on the passage, where say, know and see lead once the three are gone. Both figures are correct, and they are answers to different questions. The first asks how much a list with the helpers covers. The second asks how much twenty ordinary main verbs cover. If you are deciding which verbs to learn, you want the second.
The lab’s last tile asks the question of the passage. Count the forms of the twenty commonest irregular verbs on it and divide by all the irregular forms. With be, have and do counted, the answer is 95.3%, 142 of 149. With them left out it is 93.1%, 54 of 58. Both are counted in your browser, and both are higher than the novel’s quoted 87.66% and 73.90%. Do not read that as the passage being more concentrated. Look at the table under the tiles: with the three counted it lists 27 different verbs, and without them 24. Twenty of 27 is nearly everything, so twenty verbs cannot help covering most of the forms. A whole novel meets many more of the 133 verbs, and a long tail of them is met once or twice each. The same question gets a lower answer on a longer text. That is why the page prints its figure next to the quoted ones and does not replace them. It shows which way the number moves with the length of the text, and the novel’s figure is the better guide to the list as a whole.
Everything above comes from one novel from the 1810s. It is one writer, writing a story, with a lot of talk in it. Verbs that were common in that book may be rarer today, and the reverse. The ranks, and so the top twenty, would change in other texts. What should hold up is the shape, not the exact figures. A few verbs carry a great deal, and a long tail carries a little. The shares tell you how that shape looks in this book, and the lab lets you check it on one page of it.
That is the end of this course. The list of irregular verbs printed here is 133 long, with about fifty rarer ones left out. It falls into six patterns, and fills about one word in six of this printed page and one in eight of the novel. Three verbs carry more than half of that, and the claim that twenty verbs cover almost all of it holds only if those three are in. The Helping Verbs course takes up be, have and do next, as the words that stand in front of other verbs.
Interactive
Counted on the passage below, with and without be, have and do
| verb | times it appears |
|---|
Method
When someone gives you a share, ask these four things.
Words, or verbs, or irregular verbs only. 15.9% is of all the words in the passage. 61.1% is of the irregular forms in it. They answer different questions.
Here the question is whether be, have and do are in. They change the answer a great deal, so a figure should say.
If the largest three are most of the total, take them out and count again. The new figure shows how much the rest of the list is doing.
A figure counted in front of you on printed words can be checked by hand. A figure quoted from a count you cannot see has to be trusted, and the page should say which it is.
The first block is what the lab prints, and you can check it against the passage under the table. The second block is quoted: it was counted once over the whole novel, and the page does not carry the novel.
On the passage, 91 of the 149 forms are be, have or do: about three in five. In the novel the same three are 9,017 of 15,147 forms, which is 59.5%. The passage is one page of the book, and the book agrees with it in shape: take the three out and the share of words falls by more than half, 15.9% to 6.2% here, 12.39% to 5.01% there. The page is richer in irregular forms than the book as a whole, which is what one page of a long book can be.
The passage is where the checking happens. Count the words in the table yourself, or search the passage for had, and the tile moves when the menu does. The novel figures are the same kind of count on a text more than a hundred times longer, and they are given here as quoted. The last tile is new to this lesson. On the passage the twenty commonest verbs cover 142 of the 149 forms, and with the three left out, 54 of the 58. The novel quoted 87.66% and 73.90% for the same question, so the passage gives a higher share each time. The next paragraphs say why.
You should be able to read both shares off the lab, say how much of the total be, have and do carry, explain why a top-twenty figure that includes them flatters the list, tell a figure counted in your browser from one quoted from the whole novel, and name what one old novel cannot tell you.
Something to change about this lesson? Write it down, tick it off when it is handled, and export the list when you want to act on it. Kept in this browser only: nothing is sent anywhere, and nobody is notified.
No recommendations yet for this lesson.