When speaking English naturally and fluently, certain words shift their pronunciation depending on whether they carry sentence stress. You are almost certainly familiar with the most common example already. Articles the and a are routinely taught in their unstressed (weak) forms: /ðə/ and /ə/. What tends to get less attention is that both have strong forms too — /ðiː/ and /eɪ/ — used when the article itself is being emphasised. If you recall the main character in the movie Groundhog Day saying: “I am a god… Well, I’m not the God.”, he clearly pronounces his the as /ðiː/ as it carries semantic stress. Unfortunately, most English teaching does not go beyond the and a when it comes to weak forms. Yet this is one of the defining features of English connected speech.
What determines the weak or strong form?
Weak forms mainly occur in function words — those that hold a sentence together structurally rather than carry its main meaning — auxiliary verbs, prepositions, conjunctions, pronouns and articles. Typically in those cases the vowel reduces to schwa /ə/, but in some cases reductions may go further than that resulting in dropped sounds altogether.
The general rule is simple: a function word takes its weak form when it sits in the middle of a phrase and carries no semantic emphasis. It takes its full dictionary form when it is stressed — most naturally at the end of a phrase, or when it is being specifically highlighted or contrasted.
Watch prepositions in these two examples:
- “I did it for you” /aɪ dɪd ɪt fə ˈjuː/ — weak for
- “I did it for you, not to you” /aɪ dɪd ɪt ˈfɔː jʊ, nɒt ˈtuː jʊ/ — both prepositions stressed
While weak forms do not typically occur at the end of the sentence, personal pronouns are an exception and can go either way depending on context:
- “And you?” /ənd ˈjuː?/ — stressed
- “Are you?” /ˈɑː jʊ?/ — unstressed
The word that illustrates a particularly clean distinction. As a conjunction it almost always reduces to /ðət/. As a demonstrative pronoun it keeps its strong form /ðæt/ — even mid-sentence:
- “The day that he left.” /ðə deɪ ðət hi lɛft/ — conjunction, weak
- “I know that! That is correct.” /aɪ nəʊ ðæt! ðæt s kəˈrɛkt/ — demonstrative, strong
- “I remember that day.” /aɪ rɪˈmɛmbə ðæt deɪ/ — mid-sentence demonstrative, still strong
The word just behaves similarly. As an adverb it readily reduces to /ʤəst/, but as an adjective meaning “fair” or “righteous” it always keeps its full form /ʤʌst/:
- “Just now!” /ʤəst ˈnaʊ/ — adverb, weak
- “The just and noble men.” /ðə ˈʤʌst ənd ˈnəʊbᵊl mɛn./ — adjective, strong
Contextual variation
Some words have more than one weak form depending on what immediately precedes or follows them. The definite article the is the most familiar example — /ðə/ before consonants but /ði/ before vowels:
- “The apple and the plum.” /ði ˈæpl ənd ðə plʌm./
The preposition to also varies depending on the following sound:
- “To me and to us.” /tə mi ənd tʊ ʌs./
Auxiliary is can reduce to a barely audible /z/ or /s/ in connected speech, depending on the preceding sound:
- “He is here, is he? That is correct.” /hi z hɪə, ɪz hi? ðæt s kəˈrɛkt./
Putting it into practice
There is no need to memorise every possible scenario. The most valuable step is simply to start listening for these reductions in natural speech — they are present in virtually every sentence a native speaker produces, and once you start noticing them, a lot of previously puzzling fast speech will start to make sense.
More importantly, weak forms are not just a curiosity — they are the engine of English rhythm. English is a stress-timed language, which means its speech naturally organises itself into an alternation of stressed and unstressed syllables. It is precisely the reduction of function words to their weak forms that creates the unstressed “valleys” between the stressed content words — and the result is the characteristic iambic pulse of English: the da-DUM da-DUM pattern that underlies both everyday conversation and, not coincidentally, the blank verse of Shakespeare. When learners avoid weak forms and pronounce every word at full strength, they produce speech that sounds halting and foreign to a native ear, even if every individual sound is perfectly correct. Getting weak forms right is, in this sense, the single most impactful thing you can do for your spoken English after getting the individual phonemes in order.
How toPhonetics handles weak forms
toPhonetics supports weak form transcription via a dedicated toggle. When enabled, the tool applies the following logic: words in an unambiguously weak position are given their reduced form; words in an unambiguously stressed position get their full dictionary form; and where either reading is plausible, the tool presents both options for you to choose from (hover the mouse to view, click to select the next variant).
One exception is worth noting: the articles the and a are always transcribed in their weak forms (/ðə/ and /ə/) regardless of the toggle setting. Their strong forms (/ðiː/ and /eɪ/) are used so rarely in practice — only under emphatic contrastive stress, as in the Groundhog Day example above — that treating the weak form as the default makes sense in virtually every context. The only case where this would not apply is a completely isolated article with no surrounding text to establish context. Try putting the article on a separate line or in double-quotes — “the” — if you want to transcribe it in its dictionary form.
Running your text through toPhonetics with the weak forms option enabled, then reading the output aloud, is a very effective way to hear the iambic rhythm come to life — and to start producing it yourself.