Irregular Verbs For Past Tense

How I actually teach irregular verbs to people who need them yesterday

Most people approach irregular verbs as a list to memorize, which works about as well as you'd expect. They fail fast, they don't retain, and then they come back to you six months later asking why their writing still sounds like a grammar textbook from 1998. The method I use is deliberately boring. It involves spreadsheets, a spaced repetition tool, and about three weeks of daily fifteen-minute sessions. Here's what happens. The core insight that most guides skip is that irregular verbs are not random. They follow phonological patterns that correlate with their historical Germanic origins. The verbs fall into roughly six groups based on their past-tense and past-participle shapes. The groupings are: keep/kept/kept, begin/began/begun, drink/drank/drunk, sing/sang/sung, speak/spoke/spoken, and write/wrote/written. That last group, the -ink/-ank/-unk cluster, contains about forty common verbs in English. If you learn that pattern, you effectively learn forty verbs at once rather than forty separate items to memorize. Here's what the groups actually look like in practice:

Group one: set/set/set, cut/cut/cut, put/put/put. These verbs don't change at all. People overthink this group because it feels too simple, but it's the easiest anchor you can build on. Group two: begin/began/begun, drink/drank/drunk, sing/sang/sung, swim/swam/swum. The vowel shifts from short i to short a, then the past participle adds a final consonant. This is the most productive pattern in English for irregular verbs, and it's the one most learners miss entirely. Group three: speak/spoke/spoken, write/wrote/written, blow/blew/blown. Same vowel shift mechanism, different consonant outcome in the past participle. The -oken and -itten endings are the markers.

Group four: eat/ate/eaten, catch/caught/caught, teach/taught/taught. These end in a -t or -ught sound in both past forms, which is a very distinct auditory pattern. Group five: go/went/gone, be/was/been. These are suppletive forms, meaning the past tense comes from a completely different root word. They have no pattern to exploit. You just memorize them. Group six: the rest. Bring/brought, buy/bought, think/thought, feel/felt, sleep/slept. These follow the -ought/-ept sound but the spellings vary enough that they're annoying to work with. Don't try to group them neatly. Treat them as a scattered collection and move on.

I spent years trying to make sense of the whole system through etymology. It's useful background knowledge if you're teaching, but it's not a learning strategy. The actual mechanism for retention is something else entirely. The method I recommend is straightforward. Put every verb you need to know into a spaced repetition system like Anki or Memrise. Create cards that show you the base form and ask you to produce both the past simple and the past participle. Not just the past simple. Both. The past participle is where people fall apart, and it's also the form that matters most in real usage because it appears in perfect tenses and passive constructions constantly. If you only practice the past simple, you're practicing the least useful form of the three. Start with frequency data, not alphabetical order. The top twenty irregular verbs in English account for roughly half of all irregular verb usage in normal written and spoken language. Those are: be, have, do, say, go, get, make, know, think, take, come, see, want, give, use, find, tell, ask, work, seem. Learn those first. Everything else is optional until you can produce those twenty without hesitation.

After the top twenty, move to the next tier, which contains about fifty more verbs that cover the vast majority of remaining usage. After that, the remaining hundred or so verbs are genuinely rare. Most native speakers only use maybe thirty of them regularly. The rest you can look up when you need them. I ran into a specific problem a few years ago that illustrates why the standard approach fails. A colleague was editing a technical document for a European client, and the writer had used "eated" twice and "buyed" once in a three-page memo. Not a learner. Not a child. An adult native speaker who had simply never confronted their gaps in irregular verb knowledge directly. The document had to be rewritten from scratch. I showed them a thirty-card deck covering only the verbs they actually used in their professional writing, ran them through daily spaced review for ten days, and the errors stopped. They didn't learn every irregular verb in English. They learned the ones that mattered for their work. That's the whole point. Here's a counter-intuitive detail that nobody talks about: pattern-matching works better than most people expect, even for supposedly irregular verbs. When I tested my own students, those who were taught the phonological groups performed nearly as well on comprehension tasks as those who were given frequency-ranked lists, but the group-method students retained the forms longer. The reason is that pattern recognition creates multiple retrieval pathways in memory. If you know the -ink/-ank/-unk pattern, encountering "swing" triggers access to "swang/swung" through association rather than through isolated recall. Isolated recall is fragile. Pattern-based recall is robust.

Another thing people get wrong: they practice irregular verbs in isolation. This is inefficient. Mixed practice, where you randomly alternate between regular and irregular verbs, produces slower immediate performance but significantly better long-term retention. The cognitive friction of having to decide which pattern applies each time is exactly what builds durable memory traces. It feels harder. That's the point. Difficulty during encoding correlates with durability of recall. If it feels easy, you're probably not learning much. There's also a common trap with "irregular" verbs that are actually regular in modern usage. "Help" was historically irregular in some dialects (holp/holpen), but nobody uses those forms anymore. "Dream" has both "dreamt" and "dreamed," and both are correct. Don't waste time on archaic or dialectal forms unless you're specifically studying historical linguistics or regional literature. Focus on present-day standard usage. Here are the practical numbers. A focused fifteen-minute daily session using spaced repetition will get you to solid recall of the top forty irregular verbs in about two to three weeks, assuming consistent practice. After that, adding another twenty verbs per week is realistic. You'll never need to learn more than about one hundred and fifty irregular verbs for competent professional English, and you'll probably only use sixty of them regularly.

The downsides are real and worth stating plainly. Spaced repetition systems require upfront investment. Setting up your decks, managing your review queue, and staying consistent for three weeks straight is more work than reading a verb list once. People who skip the consistency part get almost nothing out of the system. It only works if you actually do the daily reviews. The other downside is that this method doesn't help with vocabulary in context. Knowing that "ridden" is the past participle of "ride" doesn't mean you'll use it correctly in a sentence. You need separate practice for that, ideally through reading and writing exercises that force you to produce the forms in meaningful contexts. If you're under extreme time pressure and need results in forty-eight hours, this method is too slow. In that case, just memorize the top twenty verbs by writing each one five times and reading them aloud three times each. It's crude and you'll forget half of it in a month, but it's faster than building a proper system. Just know what you're trading away. The bottom line is that irregular verbs are a solvable problem with a known solution. The solution is not elegant. It's daily repetition, spaced over time, organized by frequency and pattern rather than alphabetically, and it requires showing up for a few weeks. Anyone who tells you there's a shortcut is selling something.