Word choice in the media shapes what we feel, and what audiences come to believe about institutions. Imagine two newspapers reporting on the same new development on campus. One headline reads: “Universities Embrace AI to Enhance Learning.” The other: “AI Use Sparks Cheating Fears on Campus.” Both stories could describe the same guidelines, the same classrooms and the same students using the same tools.

Neither headline is false. AI can support learning, and it also raises genuine questions about academic integrity. Yet the first leans on words such as “embrace” and “enhance”, inviting readers to feel optimistic about progress. The second relies on “sparks”, “cheating” and “fears”, inviting suspicion and unease. The subject has not changed. The changes are reflected in the words used, and with them, the feeling. This is one of the quietest and most powerful features of media language. Words carry information, but they also carry feeling, and readers often register the feeling before they have weighed the facts.

Meaning Beyond the Dictionary

Every word has a literal meaning, its denotation, and a set of associations, its connotation. “Slim”, “thin” and “skinny” can describe the same person, but each invites a different judgement. “Frugal” and “stingy” describe the same spending habits. “Confident” and “arrogant” may describe the same behaviour. There is also a subtler layer, which linguists call semantic prosody. Some words acquire an emotional colour from the company they keep.

In everyday English, the verb “cause” appears mostly with unwelcome outcomes: we cause damage, delays and concern, but rarely cause delight. Things that “set in” tend to be unwelcome too, such as rain, decay or boredom. Readers absorb these patterns without noticing them. A writer who chooses such a word brings its emotional history into the sentence, whether intended or not.

The Linguistic Toolkit of Emotion

Media language draws on several tools to colour how readers feel, and five are especially common. The most basic is word choice. Euphemism softens, while its opposite, dysphemism, sharpens: “restructuring” feels more measured than “mass layoffs”, and “passed away” is gentler than “died”. “Freedom fighter” and “insurgent” may even describe the same person, yet invite opposite sympathies.

Individual words, however, sit within larger frames. Every story selects some aspects of reality and makes them more prominent, so a budget described as “investment” feels different from one described as “spending”, even when the figures are identical. In this way, framing shapes not only what readers feel, but what they think should be done.

Metaphor often does the work of framing, and it is far from decoration; metaphors structure how we think. Describing the economy as an “engine” suggests something that can be fuelled and sped up, while calling it “fragile” or “recovering” casts it as a patient in need of care. Likewise, describing education as a “race” can build motivation and urgency, but it can also imply that students who take longer have “fallen behind”, whereas calling it a “journey” makes room for different paces and paths.

Beyond the choice of images lies the structure of the sentence itself, because who does what to whom is also an emotional choice. Compare “The airline cancelled the flight”, “The flight was cancelled” and “The flight didn’t go ahead”: each step removes the actor a little further. Passive voice and nominalisation can thus leave anger, sympathy or blame with nowhere to land.

Finally, even when the actors and facts are clear, writers can turn the emotional volume up or down. Prices that “rise” feel different from prices that “soar”, and markets that “fall” are calmer than markets that “plunge”. Hedges such as “reportedly” or “may” work in the opposite direction, lowering certainty and, with it, alarm.

Why Negative Words Travel Further

Psychologists have long observed that bad is stronger than good. Threats demand our attention more urgently than pleasures, and news language reflects this.

A large study published in Nature Human Behaviour in 2023 put numbers to the pattern. Researchers analysed about 105,000 headline variations tested on a popular news site, covering more than 370 million impressions. For a headline of average length, each additional negative word raised the click-through rate by 2.3%. Each positive word lowered it by about 1%. Interestingly, not all negative feelings behaved alike: words linked to sadness increased clicks, while words linked to fear and joy reduced them, and anger had no clear effect.

Yet this is only half the story. Studies of everyday language across many languages suggest that people use more positive words than negative ones. Everyday language leans towards warmth; headlines competing for attention lean towards alarm. The gap between the two is where communicators make some of their most important choices.

When Negativity Accumulates

A single negative headline may nudge a reader. A cluster of them, published within days on the same theme, works differently. Each article reinforces the emotional tone of the last, and the feeling gathers weight.

The volume of coverage tells audiences which issues matter, and repeated exposure shapes how people perceive reality itself. When concentrated coverage is also worded negatively, audiences may come to feel a situation is worse, or more widespread, than the facts suggest.

There are signs that audiences push back. Global surveys show that more than four in ten people now avoid the news sometimes or often, up from fewer than three in ten in 2017. Some of the sharpest recent rises have followed intensive coverage of a single crisis or a run of scandals.

Feelings Across Languages: The Malaysian Context

In a multilingual society such as Malaysia, these choices multiply. A single story may reach readers in Bahasa Malaysia, English, Mandarin and Tamil, and the emotional weight of a word does not always survive translation.

Consider how we describe people who enter a country without documents. The official Malay term may be rendered in English as “illegal immigrants”, “undocumented migrants” or “irregular migrants”. Each points to the same group. Each also invites a different feeling, ranging from suspicion to sympathy.

Words in a speaker’s first language often carry stronger emotional resonance than equivalents in a later-learned language. This helps explain why Malaysians often switch languages mid-conversation when the subject becomes personal, humorous or heated. It also means a message crafted in English may land with a different intensity once it is shared in Malay or Mandarin.

Cultural norms matter as well. The value of budi bahasa, courteous and considerate language, shapes expectations of how criticism should be expressed. A blunt construction that seems neutral in one media tradition may feel harsh or disrespectful to a Malaysian audience. For communicators, being bilingual is not enough; we also need to be bicultural in our sense of how words feel.

What This Means for University Branding and Reputation

For universities, emotional language in the media is not an abstract concern. Much of what the public feels about an institution is formed through coverage it does not write. The tone and volume of that coverage act as a reputational asset in their own right. The qualities the media repeatedly attach to a name become the qualities the public remembers.

This is where concentration matters most. A single critical story usually fades. A burst of negatively worded articles within a short period, perhaps following an incident on campus, can fix an emotional association to a university’s name. Words repeated across outlets travel with the brand. Search results and social media feeds keep them visible long after the facts have been clarified.

The effect is not limited to the institution named. When one organisation faces a crisis, the damage can spill over to others in the same sector. A wave of stories about graduate unemployment, for example, can colour public feeling about universities generally, including those never mentioned.

Universities also shape this tone through their own words, since their statements are often quoted directly. Vague or defensive phrasing, such as “the matter is being looked into”, can deepen negative sentiment. Clear, accountable wording, such as “we identified the problem and have taken these steps”, gives follow-up stories something constructive to report.

For corporate communication teams, this suggests a few practical shifts. Firstly, track tone, not only volume. Monitor the emotional vocabulary attached to the university’s name over time, not just the number of mentions. Secondly, watch for clusters. A sudden concentration of negative wording is an early warning, even before it becomes a crisis. Thirdly, respond early, in plain language. Statements issued while a story is still forming have the best chance of shaping its tone.

The positive side is equally real. Consistent, evidence-based stories build a reservoir of goodwill over time. When negative coverage does arrive, that reservoir helps audiences read it in context rather than as confirmation of a fear.

Choosing Words Responsibly

Media language need not be emotionless, since feeling helps people understand what matters. The real question is whether the emotion a word creates is faithful to the facts. UTM’s notion of adab, right conduct that places things and people in their proper place, offers a useful guide: skill with language must be matched by care in how it is used.

For writers, this means a few habits. Choose precision over intensity, asking whether “crisis” or “soar” is accurate or merely striking. Keep the actor visible where responsibility matters. Check what feeling a metaphor carries before using it. Use people-first language, such as “people with disabilities”, which describes a circumstance rather than defining a person. Finally, write headlines that readers would still find fair after reading the full article.

For readers, media literacy starts with the same questions: which words are doing the emotional work, who the actor is, and how another outlet might describe the same event.

Generative AI makes this more urgent, as it can produce emotionally tuned headlines in seconds and reward whatever wins the most clicks. Editorial judgement, guided by adab, remains the safeguard. Emotive language may win the first click, but measured, accurate language is what earns the reader’s trust and return.

Words That Last

Return, for a moment, to the two headlines about AI use. Neither is a lie, but each asks readers to feel something different, and feelings shape what people do next: whether they listen, worry, take sides or look away.

Language specialists have always known that words do more than name things. In today’s media environment, where every headline competes for attention and AI can generate endless variations, that knowledge has become a public responsibility. Choosing words carefully is not about removing emotion from the news. It is about making sure the feelings our words create are ones the facts can honestly support.

Author

Assoc. Prof. Dr Farhana Diana Deris serves as Director of the Corporate Communications Centre at the Office of Strategy and Global Eminence, Universiti Teknologi Malaysia (UTM). She is also a Senior Lecturer at the Language Academy, Faculty of Social Sciences and Humanities. Trained in Teaching English as a Second Language (TESL), her work sits at the intersection of language, culture, and global engagement. She is particularly interested in how storytelling and ethical communication can build trust and strengthen intercultural understanding in education and society. Her research explores technology-enhanced language learning, multilingualism in tourism communities, and the internationalisation of higher education. She is actively involved in professional networks such as NEAS Quality Assurance in Education and Training, AsiaTEFL and the Malaysian English Language Teaching Association.