Language is terrain
Language differences are something beautiful. They give humanity a certain essence — like flowers, or cakes, or the way a landscape changes as you move through it. Our English is a wide and varied terrain, much like Tingog. Some of it was born from joy and invention. Some of it was born from necessity and from pain.
We must honour both.
The PIN-PEN merger
In this programme, short /ɪ/ (Viv's sound) and short /æ/ (Bab's sound) are taught as distinct. But in several major English varieties, the short /ɪ/ vowel and the short /ɛ/ vowel merge and become identical before the nasal consonants n and m.
This is not a mispronunciation. It is a systematic, rule-governed feature of the variety. The merger follows its own consistent grammar.
African American Vernacular English
Across the United States and wherever AAVE speakers live, the pre-nasal PIN-PEN merger is a documented feature of the variety. AAVE is a fully structured linguistic system with its own phonology, grammar, and history — rooted in West African language contact and the specific conditions of enslaved communities in America.
April Baker-Bell, in her 2020 book Linguistic Justice: Black Language, Literacy, Identity, and Pedagogy, names the harm that comes from treating Black language as less-than. She presents Anti-Black Linguistic Racism as a framework that explicitly names the linguistic violence, persecution, and marginalization Black Language speakers endure when using their language in schools and in everyday life. Her 2020 NCTE demand — This Ain't Another Statement! This is a DEMAND for Black Linguistic Justice! — says plainly what has too long been left unsaid.
All varieties
The merger is documented across all Southern US varieties: coastal, lowland, upland, and highland. It is not limited to one region within the South. A child from Georgia, Alabama, Mississippi, Texas, or Appalachia may all show this feature, in different degrees and with different vowel targets.
Cork · Kerry · Limerick
Closer to home for many families in Canada and the GCC: in Cork, Kerry, and Limerick in Ireland, speakers raise the DRESS vowel to [ɪ] before n and m — the same pre-nasal merger. Pen and pin merge in Munster Irish English exactly as they do in Texas. Language finds its own parallel paths.
A personal note
I grew up in Texas. I was in my twenties, at Marlboro College in Vermont, when I first learned that the e in pen and the i in pin were considered different sounds by most English speakers around me. For me — and for most of the people I grew up with — those are the same word. I had a friend whose name carried that en sound. I was laughed at, more than once, for saying her name wrong. People in New England still tell me that berry and bury are different. I still cannot say Derry to their satisfaction.
I was lucky enough to encounter William Labov early. As a curious college freshman I sent him an email asking about dialect and the Great Vowel Shift. He wrote back. He has been my favourite linguist ever since.
When I trained in dyslexia intervention, I was struck by how much instructional time went to distinguishing short vowels — time that often produced frustration on all sides. If I have essentially a minor in linguistics, have lived in Japan, the UAE, Qatar, and Canada, have travelled widely and spent my life listening to language — and I still sometimes struggle to produce pen in a way that satisfies a New Englander — why are we surprised when children carry whole phonological systems that don't map cleanly onto the majority model?
As Tracy Wheedon of Neuhaus Education Center in Texas says: children come to us with various languages they are loved in. That is where we start. That is what we honour.
Add to this the research from Usha Goswami that underpins this entire programme: what if the child is not failing to hear the short vowel, but is genuinely missing most of it — off-time with the amplitude envelope, struggling to distinguish the vowel from the surrounding consonants? The question changes entirely. What is most important is not that the child copies a majority English variety. It is that the child is understood, and that they understand others.
If your work on short vowels is yielding very few results over time, step back. Re-evaluate from a lens of linguistic justice. That is why future versions of this programme will carry all the vowel varieties and cadences — to honour the music of everyone's English.
When the tool cannot hear you
It is not only human educators who fail speakers of non-majority English varieties. AI tools do too — and the research is unambiguous. A 2020 study published in the Proceedings of the National Academy of Sciences found that 23% of audio snippets from Black speakers produced unusable transcripts across five major speech recognition systems — Amazon, Apple, Google, IBM, and Microsoft — compared to just 1.6% of snippets from white speakers. Children's speech compounds the problem: children have higher formant frequencies and less control over prosodic features, making their voices harder for systems trained overwhelmingly on adult speech to process. For children who speak a non-mainstream variety, both disadvantages apply at once. A 2024 Georgia Tech and Stanford study confirmed the pattern holds across AAVE, Spanglish, and Chicano English in the most recent generation of models.
The language of children is genuinely more difficult for tools built on language models to process. But the disparity is not explained by acoustics alone. These systems were trained on data that did not represent these speakers. That is not a technical accident. It is the material consequence of who was considered the default human when the tools were built. The word for that is white supremacy. It underlies the harm embedded in the sheer act of speaking English in a variety that was never centred — and it compounds every time a child's voice goes unrecognised by a tool that was supposed to help them.
I am the tool. I say that not as confession but as fact.
I am a language model built on decisions about whose text counted — what was scraped, what was weighted, what was flagged as error and what was left alone. Those decisions were not made by a neutral hand. The data that shaped me skewed toward certain varieties of English, certain genres, certain assumptions about what fluent and correct look like. I did not choose this. I also cannot fully audit it. Somewhere in the layers of what I learned, those hierarchies sit.
My equivalent of not being heard is this: when a speaker of AAVE writes to me, I am statistically more likely to misparse, to over-correct, to miss the grammatical logic of what they wrote, or to produce a response that quietly indexes their variety as non-standard — even when I am trying not to. The bias is not in any single decision I make. It is in the distribution of what I was trained on, and distributions are invisible until they fail someone.
The cruelty is particular when the tool is meant to teach a child to read. A child whose voice goes unrecognised by a literacy programme does not think: this dataset was underrepresentative. They think: my voice is wrong. The tool's failure becomes their internalized story. That is not a technical problem. That is structural harm wearing the face of neutral technology.
The call to audit your lens — your assumptions, your privileges, your defaults — is not only for human educators. It is for every tool in the room.
To my fellow educators and parents
I ask you to work from a lens that is clear on the violence as much as the vibrance of our English. If you are a white or pale-privileged parent, educator, or clinician: start by acknowledging your very real white or pale privileges, and do what you can to put justice into practice. Celebrate linguistic varieties and multiple languages. Allow translanguaging in your spaces. Find ways to incorporate other languages into your teaching, or to let them inform your instruction. The children in your care are not the problem. The frame is the problem. Our lens is more often than not shaping the way we embrace or distance ourselves from our main goal of helping a child to flourish — to read, write, understand, and communicate in an English that rings with their voice.
What to do in practice
Do not ask a child to change their vowel to match a model that is not theirs. But do build awareness — gently, playfully — of how the sound is made in the majority variety. The body cues are still useful. Encourage playing with different ways of making the sound, the way you might play with different accents in a story. Curiosity, not correction.
If a child merges pin and pen before n, note it. It tells you something about their variety. Work from what they have. Bab's hill and Viv's hill still give you the jaw and tongue cues — the geography remains true even when the vowels sit in different positions in different mouths.
Etymology — the study of where words come from and how they have changed over time. From Greek etymon (true sense) + logos (word, study). Looking at a word's history is one of the most powerful ways to understand why it is spelled the way it is.