# Why can't we understand text without subvocalizating it?

I wonder why can't we understand text the same way we understand pictures?

Everything we see but text we understand without the inner voice, we don't subvocalize/describe a forest, a street or anything. We just understand it.

Why can't reading work the same way? When I look at words I recognize them but they have no meaning until I read them in my mind.

Is there even anyone who got rid of subvocalization completely and can read a 300 page book just in one evening understanding pretty everything?

It even works the same the other way around. If I try to see a textural representation of some thing like let's say a table, I first have to say in my mind. Why is it so? It's really annoying.

text > voice (in my mind) > understanding


why not

text > understanding


I also don't get it why people consider this normal. It slows everything down. Imagine you could read a book in a single eventing if not the subvoc. We must do something terribly wrong when learning to read.

I'm not asking about speed-reading because subvoc is not only an issue there - although it is most noticable when reading - but it also interfers when thinking and forces you to say your thoughts.

• Do people always subvocalize? In fact, I heard that people who subvocalize texts read slower, and that faster readers do not subvocalize (or do it less often). – Robin Kramer Oct 12 '16 at 6:49
• @RobinKramer indeed subvoc slows you down because I can read only as fast as I can say the words in my mind and I can say them as fast as I can speak even if I only think I speak. I spoke to many people and even if they thought they wouldn't subvocalize they all confirmed that they were all speaking in their minds. Reading is not the only thing we subvocalize. Also when you think you do it too. If you are mad at something you don't just think you're mad and that's it, you say bad words to yourself even if no one can hear you - the same when you think of something you talk to yourself ;-/ – Octopus Oct 12 '16 at 6:56
• "we don't subvocalize/describe a forest, a street or anything" I guess we do. Because we categorize (or cognise) these things only if we named it. So categorisation (and, consequently, thinking) is probably impossible without language. And language, in turn, based on vocalization. However, I guess it tells about the current state of our approach in thinking, and not about fundamental impossibility. – user13957 Oct 13 '16 at 15:10
• @RobinKramer I suspect that the word 'subvocalize' is being used incorrectly in this question. Normally it refers to: someone moving their mouth and throat (even minutely) as if they were going to speak, but not actually making sound. But it seems that the question refers to the experience of 'hearing' words of text in one's mind while reading. – user9634 Oct 14 '16 at 14:29
• Maybe we could try teaching someone to read before they are exposed to hearing spoken language and see how they can comprehend written words? Wait... Congenitally deaf people. Ask them! – user9634 Oct 15 '16 at 20:00

In japanese language, kanjis are more than phonemes but represent an idea. For instance, both tree and spirit have the same sound (ki) and can be written with the hiragana "letter" き. But when kanjis are used tree is 木 while spirit is 気. Therefore, it is possible to read without subvocalizing. We can read as if it was graphics. So, when using kanji is much more like text > undestanding. But is hard to think doing it purely because we use many words that have no physical meaning such as "as", "such", "meaning" etc.

• This difference between writing concepts and writing sounds is indeed fundamental. I like this explanation of how they differ in experience. – user9634 Oct 14 '16 at 14:31

Visual experience is 2 dimensional, or even 3 dimensional (moving things). Spoken and heard language is one dimensional in the sense that it must be processed in order: either sounds in order or written symbols in order. The ordering is important. Visually, because things move around, there is no 'ordering' of what we see, so the brain developed ways of processing it more holistically. (Consider the well-known description of "what a frog sees".)

Even still, we understand a visual image by having the eye fixate on various points all around it (as revealed by eye-tracking studies, even with animals). And, one can see a few words at a distance and know what they say without hearing words in the mind.

Because writing is intended to stand in place of hearing spoken words, there is really no better way for us to interpret writing, other than to convert it to apparent spoken words and then process it. This is similar to how images taken in ultraviolet, infrared, X-rays, etc are converted to something we can see, and how radio waves are converted to something we can hear. It is just that we handle reading internally.

Separating the interpretation of spoken language from a simulated hearing of it would be a feat. But writing can be made distinct from hearing, if concepts (ideograms) are written instead of sounds - as in Chinese, Japanese and similar writing systems. These written languages are freed from the problems of dialect and change over time: people with many different spoken languages can all read written ideograms just fine.