Unit 1 · p.13 Family Matters

Quick confession before we start. For about ten years I believed that goldfish have a three-second memory. Turns out they can remember things for months. Who's got one like that? Something you believed and later found out was rubbish?

I used to believe that we only use ten percent of our brains, until I found out it's a myth.

Classic. Where did you get it from?

A film, I think. And then everyone repeated it.

So, here's the question on the board: do you believe everything you read on the internet?

No, of course not. But I believe more than I should, probably.

Honest. Now the harder one: what if it's based on scientific research? Does "studies show" make you trust it more?

Yes, because a research is more reliable than an opinion.

In principle, yes. Small grammar point first: research is uncountable, so no "a". We say "a study" or "a piece of research". Then tell me: is every study reliable?

A study is usually more reliable than an opinion, but it depends on how it was done.

"Depends on how it was done." Let's make that concrete. Think back to "The Power of Siblings". Which claim convinced you most?

The Harvard one. They followed three hundred men for their whole lives. That's a long time.

Long time, decent number. Anything missing from that study?

Oh. Only men. So we don't know about women.

Nice catch, that's exactly the kind of question the Train to Think box wants. And which claim convinced you least?

The one that siblings make you physically healthier. It just says "there's evidence", but not whose evidence.

So you trust a claim more when you can see who, how many, and how long. Remember the girl from last lesson who said "only children are spoiled, everybody says so"? That's the opposite: no who, no how many, just "everybody".

Today's question: if a study says one thing, and the people you actually know say another, how many exceptions does it take before you stop believing the study?

Look at the red box in the bottom left: Train to Think, "Questioning widely accepted theories". Read it in silence, thirty seconds.

Where do conclusions come from, according to the writer?

From what we've read, heard or observed.

So when we make a conclusion from what we observed, it can be wrong.

Right idea. But conclusions are drawn, not made. Draw a conclusion. Like drawing water from a well: you pull it out of the evidence.

So when we draw a conclusion from what we've observed, it isn't always a hundred percent correct.

Now careful: not a hundred percent correct. Does that mean wrong?

No. It means it's true most of the time, but not every time.

That's the key distinction. Now three words that people mix up: finding, claim, theory. The Norway study found that older siblings had higher IQ scores. Finding, claim or theory?

Finding. It's what the study actually discovered.

Now the heading says "Siblings can make you smarter".

That's a claim. It goes further than the finding.

How much further?

The finding is only about older siblings in Norway, but the heading talks about everyone, and it says siblings MAKE you smarter, as if it's the cause.

That's a big jump, from "is connected with" to "causes". And "first-born children have all their parents' attention"?

That's a theory. It tries to explain why.

Perfect. So: finding, claim, theory. Each step adds a little more guessing. Last line of the box: "There will always be plenty of exceptions." If exceptions are guaranteed, why bother with research?

Because it still shows a pattern for big groups, even if it can't predict one person.

For example, the research can't say me will be smart.

Exactly the point, wrong pronoun. After "say" we need a full clause with a subject: it can't say that I'll be smart.

The research can't say that I'll be smart, but it can say that older siblings, on average, tend to score a bit higher.

"On average" — that's the phrase scientists love and headlines delete.

Back into the article. Not to understand it this time, but to cross-examine it. For each heading, find the evidence: who, how many, where, when.

两人一组，每组负责两到三个标题，3 分钟。

Smarter. What's the evidence?

A 2007 study in Norway. Older siblings tend to have a higher IQ.

And how confident is the language?

Quite careful. "Tend to", "might be due to", "could be a result of".

Good, so there's a named study and hedged language. Now empathy.

It just says "there's evidence to suggest". No study, no country, nothing.

And then what does the writer say?

"These traits, of course, carry over into adulthood", and "it's known that empathy is a determining factor in adult happiness".

So the claim with the weakest evidence has the most confident language. "Of course". Says who?

The writer is too confident, it makes me doubt about the claim.

That's a really sharp observation. Grammar: doubt is a verb on its own, no "about". It makes me doubt the claim. Or as a noun: it makes me have doubts about the claim.

The writer sounds too confident, which makes me doubt the claim.

Mental health: Harvard, around three hundred men. How strong?

Fairly convincing, since they followed them for a long time. What the study doesn't tell us is anything about women.

And Ohio State, the divorce statistic?

"Your chance of getting divorced will fall by as much as two percent." But "as much as" means that's the maximum. It could be much less.

Brilliant reading. "As much as two percent" sounds big, but it's a ceiling, not an average. Now one cultural detail about IQ. It's mental age divided by real age, times a hundred, from standardised tests like the Wechsler scales. Two-thirds of people land between 85 and 115.

But IQ is linked to things like family income, nutrition, even conditions before birth. So why is it always an estimate?

Because intelligence is abstract. You can't measure it directly, only test some parts of it.

So even the measuring tape is a bit wobbly. Rate the five claims from one to five, and be ready to defend your strongest and weakest.

I gave Harvard a four, since it's long-term, and empathy a one, since there's no source at all.

Exercise 6. First, the key word: contradict. If two statements contradict each other, what's going on?

They can't both be true. One says the opposite of the other.

So example 1 is contradict with the article.

You've spotted it, I can tell. But two fixes. Contradict is already a verb, so no "is". And it doesn't need "with": A contradicts B.

So example 1 contradicts the article.

Which claim, exactly? Point to it.

"Older siblings tend to have a higher IQ than younger ones." But Tom is the youngest and he's way smarter than Andy.

And there's a second, sneakier contradiction in example 1. Look at who helps whom.

Tom helps Andy with homework. But the article says smart older siblings encourage the younger ones. Here it's the other way round.

Double contradiction. Now example 3, Steve.

The article says poor sibling relationships seem to produce adults who are insecure and sometimes depressed. Steve wasn't close to his brothers, but he's always upbeat and positive. So it contradicts it.

Now the tricky one. Example 2. Mary, the eldest of four, has the fewest friends. Vote: contradicts, or doesn't? Hands up.

全班举手投票，记下两边人数。

Contradicts. The article says the more siblings, the better at managing relationships. Mary has three siblings, so she should be good with people.

Interesting argument. Somebody on the other side?

I disagree. The article compares families with more or fewer siblings. Mary's brothers and sisters have exactly the same number of siblings as she does. So the article can't say anything about who has more friends inside one family.

Ooh. That's a really important distinction. Also, is "fewest friends" the same as "bad at managing relationships"?

No. She might be good with people but just choose a few close friends.

So the answer is 1 and 3. Example 2 sounds like a contradiction, but it's about something the article never actually claimed. That's the trap: before you say "this contradicts it", find the exact sentence it contradicts.

One more question. Tom contradicts the Norway study. Does that make the study wrong?

No, the study says "tend to". One exception doesn't break a tendency.

Exercise 7. Now it's personal. One minute: write down four or five people you know and where they are in their family. Eldest, middle, youngest, only child.

学生写笔记 1 分钟，同时给每个人想两个形容词。

Next to each name, two adjectives. Honest ones.

Now find someone who breaks one of the article's claims. Tell your partner who they are, which claim, and how.

My uncle is the youngest of three, and yet he's the most empathetic person in the family.

And what does the article say?

It says younger siblings teach the older ones empathy. So if the article were right, you'd expect his older siblings to be the empathetic ones. But in fact, they tend to be quite self-centred.

Lovely use of "if the article were right". Is there another explanation?

My guess is that it has more to do with his job. He's a nurse. He's always looking after people.

My cousin is the only child, but she is very easy to share things.

I love that example, because it's exactly what the stereotype says she shouldn't be. Grammar check: a person isn't "easy to share". She's generous, or she's always happy to share things.

My cousin's an only child, but she's incredibly generous. She's always happy to share things.

And the article's claim?

It says only children can be more self-centred, because they don't have younger siblings to teach them empathy.

Other explanation?

Her parents made her share her toys with all the neighbours' kids. She grew up with lots of children around.

So she was an only child on paper, but not in practice. That's an important point. "Only child" might not even be the right category.

My brother is the eldest, so according to the article he must be smarter than me. It's not true!

Ha, family politics! But check the word "must". Does the article say "must"?

No, it says older siblings tend to have a higher IQ. So my brother doesn't contradict it really. I'm just one exception.

Exactly. So: how many exceptions have we found in this room in five minutes?

At least six or seven.

And does that make the research wrong?

Not wrong. But it makes it less useful for predicting one person.

Exercise 8. Beyond this article, what other theories about families have you heard? Let me get you started.

The middle child is always ignored. Middle child syndrome.

A famous one. Where did you hear it?

From TV shows mostly. And my middle cousin says it all the time.

The middle child is easy to be ignored by the parents.

That's a very Chinese sentence structure, 容易被忽视. In English we can't say a person "is easy to be ignored". Try "tends to be overlooked" or "often gets overlooked".

The middle child tends to be overlooked by the parents.

Better. Another one, maybe a Chinese one?

长兄如父. The eldest brother is like a father to the younger ones.

Nice. How would you put that in English as a theory?

There's a widely held belief in China that the eldest son should act like a second father.

Good. Is that a finding from research, or a saying?

A saying. It's more about tradition than about evidence.

Two minutes. In pairs, list three theories and one person who contradicts each.

讨论 2 分钟后，打乱重组，新组之间交换信息。

New groups. Share your best theory and your best contradiction.

There's a widely held belief that the youngest is spoiled. It doesn't hold for my dad, who's the youngest of five. He had to wear his brothers' old clothes his whole childhood.

Ha, so the youngest of five in the eighties: very spoiled. Why do theories like this survive, even though everyone knows exceptions?

I suspect it survives because it's simple and easy to remember.

And when someone fits the theory, we notice it. When they don't, we forget.

That's confirmation bias, beautifully described. We notice the cases that fit and ignore the rest. So back to the question: how many exceptions does it take?

Last six minutes. Our question: how many exceptions does it take before a scientific finding stops being true for you? Write three or four sentences.

Structure: one claim from the article, one real exception, your conclusion. Checklist on the board.

写 3 分钟，然后同桌交换检查。

The article claims that older siblings tend to have a higher IQ. My mum's family contradicts this, because her younger sister is the cleverest. But one exception doesn't make a finding false, because it only describes what happens on average.

Strong. And your conclusion: how many exceptions?

For me, a finding stops being useful, not true, when it's used to predict one particular person.

"Useful, not true." That's a really mature distinction. Check your partner's: "contradicts" without "with"?

He wrote "my cousin contradicts with the claim".

And what's the fix?

Delete "with". My cousin contradicts the claim.

Now homework. You're going to become a pseudo-scientist. Invent your own personality theory. For example: people born in January tend to be self-centred.

People born in July are always organising other people!

Exactly that energy. Write your theory, and next lesson you'll test it on at least five classmates and report back. Then we'll see how many exceptions your theory can survive.

Today you learned to separate a finding from a claim, to spot confident language with weak evidence, and to find exceptions without throwing the research away. Next time: grammar for talking about habits, including the annoying ones.