TRANSCRIÇÃO 🔗
WEBVTT
1
00:00:00.150 --> 00:00:03.770
You explained it carefully. You chose your words. You went slowly.
2
00:00:04.389 --> 00:00:07.230
You carefully explained it. You checked whether they were following.
3
00:00:07.889 --> 00:00:10.410
And at the end, they nodded. And they repeated back
4
00:00:10.470 --> 00:00:13.919
something completely different from what you said. Not the opposite,
5
00:00:14.279 --> 00:00:17.300
just different. As if your explanation had gone into their
6
00:00:17.339 --> 00:00:21.100
head and come out rearranged. And you were standing there thinking,
7
00:00:21.620 --> 00:00:25.039
how is that possible? I was so clear. Welcome to
8
00:00:25.120 --> 00:00:28.760
Before You Overthink It Again. I am Logan. Today we
9
00:00:28.800 --> 00:00:31.359
are talking about the curse of knowledge, and about why
10
00:00:31.420 --> 00:00:34.619
being clear is not the same as being understood. Here
11
00:00:34.659 --> 00:00:38.280
is the question. Why does explaining something clearly still end
12
00:00:38.320 --> 00:00:42.280
in misunderstanding? You did everything right, you were patient, you
13
00:00:42.320 --> 00:00:45.990
were precise, you used plain language, and the other person
14
00:00:46.070 --> 00:00:49.590
still heard something else. Did they not listen? Did you
15
00:00:49.630 --> 00:00:52.590
not explain well enough? Or is something else going on,
16
00:00:53.109 --> 00:00:55.670
something that has nothing to do with effort on either side?
17
00:00:56.270 --> 00:00:58.780
Something else is going and it has a name worth knowing,
18
00:00:59.259 --> 00:01:02.310
the curse of knowledge. It was demonstrated in a study
19
00:01:02.369 --> 00:01:06.030
by Kammerer, Lowenstein, and Weber in the late 1980s, and
20
00:01:06.049 --> 00:01:09.430
it describes a very simple problem. Once you know something,
21
00:01:09.689 --> 00:01:12.489
it is hard to imagine not knowing it. The knowledge
22
00:01:12.549 --> 00:01:14.849
sits in your head, and it colors everything you say
23
00:01:14.909 --> 00:01:17.219
about the topic, and you cannot take it out again,
24
00:01:17.540 --> 00:01:21.299
even when you try. You explain from inside your own understanding,
25
00:01:21.780 --> 00:01:24.859
skipping steps that feel obvious to you, using words that
26
00:01:24.920 --> 00:01:27.530
carry a full meaning for you, and assuming the other
27
00:01:27.569 --> 00:01:30.390
person is starting from the same place you are. They
28
00:01:30.430 --> 00:01:33.989
are not. They are starting from somewhere else entirely, and
29
00:01:34.010 --> 00:01:36.230
the gap between the two starting points is where the
30
00:01:36.250 --> 00:01:40.069
misunderstanding lives. There is a famous demonstration of this, and
31
00:01:40.090 --> 00:01:42.209
it is worth hearing in full, because it is almost
32
00:01:42.329 --> 00:01:47.109
comically clear. In 1990, a researcher named Elizabeth Newton ran
33
00:01:47.159 --> 00:01:51.019
an experiment with tappers and listeners. One person, the tapper,
34
00:01:51.480 --> 00:01:55.019
was given a well-known tune, something like Happy Birthday, and
35
00:01:55.120 --> 00:01:57.870
asked to tap it out on a table. The other person,
36
00:01:58.090 --> 00:02:00.629
the listener, had to guess the tune from the tapping.
37
00:02:01.430 --> 00:02:04.209
Before they started, the tappers were asked to predict how
38
00:02:04.329 --> 00:02:07.750
often the listeners would guess correctly. The tappers guessed about
39
00:02:07.810 --> 00:02:11.430
half the time. In fact, the listeners guessed correctly only
40
00:02:11.469 --> 00:02:15.129
about one time in 40. The tappers were not being arrogant.
41
00:02:15.610 --> 00:02:18.180
They were hearing the tune in their heads, full and rich,
42
00:02:18.379 --> 00:02:20.800
while they tapped, and they could not imagine what the
43
00:02:20.840 --> 00:02:23.280
tapping sounded like to someone who did not already know
44
00:02:23.319 --> 00:02:27.159
the song. To the tapper, the tapping was obviously the tune.
45
00:02:27.960 --> 00:02:30.340
To the listener, it was a series of knocks with
46
00:02:30.419 --> 00:02:32.900
no melody in them at all. The knowledge of the
47
00:02:32.939 --> 00:02:35.280
tune made it impossible to hear the tapping the way
48
00:02:35.319 --> 00:02:38.629
the listener heard it. And that is the curse. You
49
00:02:38.650 --> 00:02:42.270
cannot unhear what you know. Now apply that to your explanation.
50
00:02:42.990 --> 00:02:46.050
You know the background. You know the context. You know
51
00:02:46.129 --> 00:02:49.069
why this matters and what the words mean. And as
52
00:02:49.110 --> 00:02:51.509
you explain, you hear all of that in your head,
53
00:02:52.009 --> 00:02:54.759
the way the tapper hears the tune. The other person
54
00:02:54.800 --> 00:02:57.939
hears only the tapping. They hear the words. But they
55
00:02:57.960 --> 00:03:00.400
do not hear the background, because the background is in
56
00:03:00.419 --> 00:03:03.590
your head, not in the words. And the more expert
57
00:03:03.629 --> 00:03:06.550
you are, the worse this gets, because the more you know,
58
00:03:06.930 --> 00:03:09.229
the more background there is, and the harder it is
59
00:03:09.270 --> 00:03:11.219
to remember what it was like not to have it.
60
00:03:11.969 --> 00:03:14.669
The people who are worst at explaining something are often
61
00:03:14.689 --> 00:03:17.259
the people who know it best. Not because they are
62
00:03:17.280 --> 00:03:20.300
bad teachers, because they are cursed by their own knowledge.
63
00:03:20.939 --> 00:03:22.919
There is a second layer here, and it is on
64
00:03:22.939 --> 00:03:27.169
the listener's side. Listeners nod. They nod because they recognize
65
00:03:27.210 --> 00:03:30.449
the words, not because they follow the logic. They nod
66
00:03:30.490 --> 00:03:33.870
because nodding is polite, and because saying, I do not understand,
67
00:03:33.969 --> 00:03:37.189
feels like
Load full transcript
an admission, and because the explainer seems confident,
68
00:03:37.550 --> 00:03:41.189
and confidence is persuasive. So the explainer sees the nodding
69
00:03:41.460 --> 00:03:45.159
and concludes the explanation worked, and the listener, having nodded,
70
00:03:45.680 --> 00:03:47.520
now has to live with a version of the idea
71
00:03:47.680 --> 00:03:50.199
that is not quite the idea, and the next time
72
00:03:50.219 --> 00:03:53.180
they use it, they will use the rearranged version, and
73
00:03:53.219 --> 00:03:56.599
the rearrangement will travel further, until no one remembers what
74
00:03:56.620 --> 00:04:00.639
the original... The nod is the enemy of understanding, and
75
00:04:00.659 --> 00:04:03.319
it is also the most common sound in any conversation.
76
00:04:03.900 --> 00:04:06.500
And there is a third layer, which is about words themselves.
77
00:04:07.189 --> 00:04:09.689
In arguments, two people will use the same word and
78
00:04:09.710 --> 00:04:14.789
mean completely different things by it.
79
00:04:14.830 --> 00:04:19.790
80
00:04:19.829 --> 00:04:24.589
81
00:04:24.629 --> 00:04:29.170
82
00:04:29.209 --> 00:04:33.800
83
00:04:33.839 --> 00:04:38.519
84
00:04:38.560 --> 00:04:43.300
85
00:04:43.339 --> 00:04:48.079
86
00:04:48.120 --> 00:04:51.850
It is a broken word phrase
87
00:04:51.930 --> 00:04:55.269
based on Japanese word, which is intended to mean great word.
88
00:04:55.959 --> 00:04:58.500
It is a word phrase used in English accent to
89
00:04:58.540 --> 00:05:02.040
describe something briefly brashed. We simply want to know what
90
00:05:02.120 --> 00:05:04.759
words we're talking about. We just want to know what
91
00:05:04.839 --> 00:05:07.819
words we're talking about. We don't care about words if
92
00:05:07.860 --> 00:05:10.819
we care about words working. The twist is a broken
93
00:05:10.879 --> 00:05:14.600
word phrase referred to a relationship of words. Every time
94
00:05:14.639 --> 00:05:16.980
we talk about words, we wonder if we know what
95
00:05:17.079 --> 00:05:19.839
words we're talking about. We just want to know what
96
00:05:19.920 --> 00:05:22.990
words we're talking about. We don't care about words if
97
00:05:23.000 --> 00:05:26.060
we care about words working. The twist is a broken
98
00:05:26.100 --> 00:05:30.810
word phrase referred to a relationship commitment. Each person hears
99
00:05:30.829 --> 00:05:33.189
their own meaning in the word and assumes the other
100
00:05:33.230 --> 00:05:35.709
person hears the same meaning, and they argue for an
101
00:05:35.790 --> 00:05:38.850
hour about a word they are not actually sharing. The
102
00:05:38.889 --> 00:05:41.649
curse of knowledge is not only about facts. It is
103
00:05:41.689 --> 00:05:45.310
about meanings, and meanings are private until someone checks them.
104
00:05:45.949 --> 00:05:49.300
So what do people commonly misunderstand here? They think that
105
00:05:49.339 --> 00:05:51.519
if they explain it well enough, the other person will
106
00:05:51.540 --> 00:05:55.569
get it. But understanding is not transmitted. It is constructed.
107
00:05:56.290 --> 00:05:59.350
The listener builds their understanding from their own background, their
108
00:05:59.430 --> 00:06:03.089
own experience, their own starting point. You cannot pour your
109
00:06:03.129 --> 00:06:06.129
understanding into their head. You can only help them build
110
00:06:06.170 --> 00:06:09.569
their own. And there is a second misunderstanding, which is
111
00:06:09.600 --> 00:06:12.720
that the other person was not listening. Often they were
112
00:06:12.759 --> 00:06:15.939
listening very carefully. They were just building a different building
113
00:06:16.000 --> 00:06:19.459
from the same bricks because their bricks were different. The
114
00:06:19.500 --> 00:06:22.579
misunderstanding is not a failure of attention. It is a
115
00:06:22.660 --> 00:06:26.019
failure of shared ground. There is a distinction worth making
116
00:06:26.120 --> 00:06:29.990
between clarity and shared ground. Clarity is about your words.
117
00:06:30.560 --> 00:06:33.269
Shared ground is about where you start. You can be
118
00:06:33.329 --> 00:06:36.410
perfectly clear and still start from the wrong place, if
119
00:06:36.449 --> 00:06:38.709
you start from what you know instead of what they know.
120
00:06:39.470 --> 00:06:42.470
The question that builds shared ground is not, do you understand?
121
00:06:42.949 --> 00:06:46.509
Because everyone says yes to that. The question is, what
122
00:06:46.529 --> 00:06:49.600
do you already know about this? Or, could you explain
123
00:06:49.639 --> 00:06:52.100
it back to me in your own words? The second
124
00:06:52.139 --> 00:06:54.699
question is the real test. because it shows you what
125
00:06:54.720 --> 00:06:58.060
they actually built, not what they nodded at. And there
126
00:06:58.100 --> 00:07:00.980
is a second part to the distinction, which is about examples.
127
00:07:01.680 --> 00:07:06.000
Concrete examples before abstract terms. Show them the thing, and
128
00:07:06.019 --> 00:07:09.000
then name it, not the other way around. A name
129
00:07:09.040 --> 00:07:12.189
without an example is just a sound. An example without
130
00:07:12.230 --> 00:07:14.990
a name is a story they can hold onto. There
131
00:07:15.050 --> 00:07:17.490
is a practical step here, and it starts before you
132
00:07:17.529 --> 00:07:21.370
explain anything. Ask where they are starting from. What do
133
00:07:21.389 --> 00:07:24.819
you already know about this? What have you heard? What
134
00:07:24.839 --> 00:07:28.139
does this word mean to you? Those three questions take
135
00:07:28.139 --> 00:07:31.000
30 seconds, and they save you from explaining the wrong
136
00:07:31.079 --> 00:07:34.399
thing to the wrong starting point. And then, while you explain,
137
00:07:34.920 --> 00:07:37.730
watch for the nod that comes too fast. A nod
138
00:07:37.750 --> 00:07:40.730
that arrives before you have finished the sentence is not understanding.
139
00:07:41.250 --> 00:07:45.129
It is politeness. Slow down. Ask them to say it back.
140
00:07:45.829 --> 00:07:48.009
And if what comes back is different from what you said,
141
00:07:48.490 --> 00:07:52.230
do not repeat yourself louder. Repeat yourself from their starting point.
142
00:07:52.670 --> 00:07:56.050
with their words, with an example from their world. Here's
143
00:07:56.089 --> 00:07:58.069
the sentence I want you to keep from this episode,
144
00:07:58.310 --> 00:08:00.209
because it is short enough to use the next time
145
00:08:00.250 --> 00:08:04.180
an explanation goes sideways. You cannot pour your understanding into
146
00:08:04.209 --> 00:08:07.160
someone else's head. You can only help them build their own.
147
00:08:07.860 --> 00:08:11.839
The building takes time. It takes questions. It takes examples.
148
00:08:12.519 --> 00:08:14.639
And it takes the humility to remember that what is
149
00:08:14.800 --> 00:08:18.500
obvious to you is not obvious to everyone. Not because
150
00:08:18.519 --> 00:08:21.379
they are slow, but because they have not lived inside
151
00:08:21.399 --> 00:08:24.740
your knowledge the way you have. So, back to the explanation.
152
00:08:25.300 --> 00:08:28.579
You were careful. You were slow. You chose your words.
153
00:08:29.230 --> 00:08:32.490
And they nodded, and they repeated back something different, and
154
00:08:32.529 --> 00:08:35.669
you were wondering how that is possible. It is possible
155
00:08:35.730 --> 00:08:38.429
because you were hearing the tune while you tapped, and
156
00:08:38.450 --> 00:08:41.090
they were hearing only the tapping. That is not a
157
00:08:41.190 --> 00:08:44.049
failure of clarity, and it is not a failure of listening.
158
00:08:44.730 --> 00:08:47.509
It is the curse of knowledge, doing its ordinary work.
159
00:08:48.360 --> 00:08:52.080
Next time, start from their side. Ask what they already know.
160
00:08:52.860 --> 00:08:55.320
Ask them to say it back, and build the bridge
161
00:08:55.360 --> 00:08:57.860
from where they are standing, not from where you are.
162
00:08:58.639 --> 00:09:02.029
The understanding will still take time, but this time, it
163
00:09:02.049 --> 00:09:05.009
will be built on shared ground, and shared ground is
164
00:09:05.049 --> 00:09:06.330
the only ground that holds.