When Two Vowels Go Walking: Is This Rule Even True?

A Rhyme Every Parent Has Heard

You've probably heard some version of the old classroom rhyme: two vowels go walking, and the first one supposedly does the talking, saying its own long name. If you grew up in an American classroom, you may have already caught yourself repeating it while pointing at boat or rain.

Here's the uncomfortable part: researchers have actually tested this rule against real words, and it does not hold up nearly as well as the rhyme suggests.

Why Researchers Bothered Checking

Reading researcher Theodore Clymer wondered the same thing decades ago. In 1963, he took 45 commonly taught phonics rules - including this one - and checked each against a list of roughly 2,600 words children actually meet in early reading materials (Clymer, 1963). His bar for a "useful" rule was generous: if it held true at least 75% of the time, he called it reliable.

The vowel-team rule didn't come close. Clymer found it worked for less than half of the words with two vowels together. A later study aimed squarely at this one rule got an even lower number: checking the same generalization against 506 real words with adjacent vowels, only about a third followed it (Bailey, 1968).

What the Rule Actually Gets Right

This is where honesty matters more than a tidy story. The rule isn't uniformly unreliable - it just isn't reliable as one blanket statement covering every vowel pair.

When researchers broke the results down by specific pair instead of lumping every vowel-team word together, four combinations did noticeably better than the rest: ai, ea, ee, and oa each worked more than half the time, some considerably more (Bailey, 1968). Words like rain, seat, tree, and boat really do tend to follow the pattern your child was taught. The trouble comes from combinations like oo, and from ea in its other job (as in bread), which behave nothing like the rule promises.

So the rule works reasonably well for a specific handful of common spellings, and works poorly as a general law for "any two vowels together." Both things are true at once, and a rule that's only sometimes right, taught as if it's always right, is exactly the kind of thing a child has to unlearn the first time it fails on a word like does or said.

Why English Isn't as Chaotic as This Makes It Look

It's worth zooming out, because one unreliable rule can make English spelling feel hopeless, and that's not a fair read either. A large-scale analysis of more than 17,000 words found that about half of them can be spelled correctly from sound alone, and another 36% are accurate except for a single sound (Hanna et al., 1966). Later estimates suggest only around 4% of words in print are truly irregular from top to bottom (Moats, 2020). Two vowels going for a walk is one specific pattern that happens to be messier than it looks - it isn't proof that the whole system is a guessing game. We go into this in more depth in English is more regular than you think.

So What Should You Teach Instead?

The most useful shift is small: teach vowel teams one pair at a time, as their own specific pattern, rather than one rule meant to cover all of them. Ai behaves like ai behaves. Oa behaves like oa behaves. Your child doesn't need a unifying theory of "two vowels together" - they need enough exposure to each common pair that it becomes automatic on sight, the way it works for ai and ay taught as its own pattern. Sweet Phonics builds its early vowel-team lessons the same way: ee, ea, ai/ay, and oa/oe each get their own separate lesson, back to back, rather than one lesson that tries to teach "vowel teams" as a single catch-all rule.

The second shift is about what to do when a pattern doesn't work: flex. If a child tries the long-vowel sound on a vowel pair and it doesn't make a real word, the useful habit isn't guessing wildly - it's trying another reasonable sound and checking whether that one makes a word they know. That's the same check-it-against-a-real-word habit that matters for open and closed syllables: a rule is a first guess, not a guarantee, and the real skill is knowing what to do when the first guess is wrong.

A Note on Why This Rule Got Taught So Widely

If the rule is this unreliable, it's fair to wonder why it became a classroom staple in the first place. Part of the answer is that it's genuinely useful for a beginner's first exposure to vowel teams: ai, ea, ee, and oa are among the very first vowel-team spellings most children meet in early books, so a rule that works well for exactly those pairs feels airtight for months. The trouble shows up later, once a child's reading vocabulary expands into words with oo, ei, ie, and other, less obliging combinations, and the rule that felt like a law turns out to have been describing a handful of specific spellings all along.

How to Practice at Home

• Teach pairs, not the umbrella rule. When you introduce ai or oa, practice a cluster of real words with that exact pattern (rain, pain, train; boat, coat, road) rather than leaning on "two vowels go walking" as the explanation.
• Collect counterexamples together, playfully. Make a short list of words where the "first vowel talks" idea fails - said, does, bread, chief - and treat it like a scavenger hunt rather than a trap. It normalizes the idea that not every pattern is airtight.
• Practice the flex, out loud. When your child hits an unfamiliar vowel-team word, ask, "Does that sound like a real word? If not, what's another way to say it?" That question matters more than any single rule.
• Don't retire the rhyme entirely. It's a reasonable first guess for ai, ea, ee, and oa specifically - just don't present it as a law that covers every vowel pair your child will ever meet.

The Takeaway

"Two vowels go walking" is a rule of thumb, not a law: Clymer's original research found it held for under half of the words tested, and a closer look found some vowel pairs - ai, ea, ee, oa - working far more often than others. The honest version isn't "this rule is wrong" or "this rule is right." It's that a handful of common vowel teams are genuinely reliable, the rest aren't, and a child who learns to flex when a pattern fails is better equipped than one who only memorized a rhyme.