We’re Not Cold-blooded Thinking Machines

I’ve just started reading Louder Than Words: The New Science of How the Mind Makes Meaning. Since my husband Jimmy has absolutely devoured a ton of linguist books over the past year I asked him to suggest one for me to read. Louder Than Words by Benjamin K. Bergen was the book he chose so off I go.

Though my interests are not specifically linguistic at this point I am still very interested in how our brains work, how it is we learn and think, and especially how it is we make meaning. So far I’ve only read the epilogue but the following quotes from George Lakoff give you a hint of what the book is about:

They [Merleau-Ponty and Dewey] argued that—quite to the contrary of the traditional view—our bodies have absolutely everything to do with our minds. Our brains evolved to allow our bodies to function in the world, and it is that embodied engagement with the world, the physical, social, and intellectual world, that makes our concepts and language meaningful. (ix)

 
The Embodiment Revolution has shown that our essential humanness, our ability to think and use language, is wholly a product of our physical bodies and brains. The way our mind works, from the nature of our thoughts to the way we understand meaning in language, is inextricably tied to our bodies—how we perceive and feel and act in the world. We’re not cold-blooded thinking machines. Our physiology provides the concepts for our philosophy. (x)

Meaning is a slippery concept… how do we actually make meaning? And how does the making of meaning affect how we understand texts? Ancient texts?? Inspired texts???

And what does having bodies have to do with it all?

The man who hunts ducks out on the weekends.

In my last post I introduced construction grammar with the help of Benjamin Bergen and his book Louder than Words. Bergen’s book, as a reminder, is an introduction to the way humans process language. In this post I want to jump to the next chapter in Bergen’s work (chapter 6) where he discusses the cause and effect of real time language processing.

Did you know that you and I can only take in language one piece of information (syllable, letter, word, etc.) at a time? I guess I knew this–it’s an observable fact. But, I never really thought about it until I started reading up on Information Structure.  Information Structure, or IS, is the interaction of pragmatics and syntax. Bergen’s work deals, not with IS, but with processes that our brains go through as we encounter language in real time.

The big idea is that there are limitations on our ability to process language due to our uptake capacity. Whenever we read or hear language we can only take it in as the string of words and sentences that it is. Our eyes have to pass over each word on down the line as we read, and our ears have to hear each syllable as it is spoken. There’s no way to mass download language. Say it ain’t so, Morpheus! This means we hear some words before others and that we can, potentially, read a bunch of words before we get the complete idea of a sentence.

The way our brains deal with this limitation is to try and figure out the whole sentence as soon as we start getting the pieces. We don’t wait to get all the words of a sentence and then process it. We process “incrementally,” making best guesses, and updating as we get more information. The process is like informed guess work where we start off with very little, but continually update as we take in–that is, read or hear–the string of language.

Since we put together sentences incrementally by making informed guesses, we have the ability to make mistakes and have to reevaluate language as we get more input. Bergen provides a few example sentences, designed to be unpredictable, called “garden path sentences” which prove this point.

(1) Time flies like an arrow. Fruit flies like a banana.
(2) The lawyer cross-examined by the prosecutor confessed.
(3) The horse raced past the barn fell.
(4) The old man the boat.

These examples trip us up because there are pieces that we analyze one way (‘raced’ in (3) as a verb) which end up needing to be reanalyzed (the verb in (3) is ‘fell’… ‘raced’ is a passive participle modifying the noun ‘horse’). Because we guess at what a sentence will be as we encounter each word, we build expectations of what the entire sentence will be as we process it. When we run into something that doesn’t match what we expect to come next, if it doesn’t fit the guessed pattern, we have to go back and reevaluate everything that we have taken in.

This phenomena is not limited to English, or to modern languages. It even happens in scripture.  I would like to look at an example from the book of James which uses this very phenomena on purpose.

James 1:2
Πᾶσαν χαρὰν ἡγήσασθε, ἀδελφοί μου, ὅταν πειρασμοῖς περιπέσητε ποικίλοις,

Consider it all joy, my siblings, when you encounter various trials

The word order of the original greek is extremely important here (as if it’s not important everywhere!). In fact, the word order is actually what creates the effect that I want to look at. Now, I am not talking about information structure. I only want to look at the expectations that are created by incremental processing. The very first words that we encounter are πᾶσαν χαρὰν (all/complete joy) which is followed by the verb ἡγήσασθε (BDAG sense 2: to think/consider). Next is a phrase directed directly to the audience (ἀδελφοί μου my siblings). Last of all is the subordinate clause ὅταν πειρασμοῖς περιπέσητε ποικίλοις (whenever you fall among various trials).

The noun phrase πᾶσαν χαρὰν (all/complete joy) is first in the sentence, and that doesn’t provide a whole lot to go off of for a reader or hearer. Being a Greek reader–I am assuming the intended readers of James held fluency in Koine and could read Greek in the same manner that you and I can read English–the reader’s mind is constrained to start filling in some of the empty information. The reader knows that this is probably the object of a verb since it is in the accusative case. So, somebody or something is doing some action where complete joy is the direct object.

Next comes ἡγήσασθε. Now the reader knows who is doing the action (s/he is), and what is being done (s/he is being commanded to ‘consider’ something). ἡγέομαι is a verb that takes two accusative nouns. In my last post I used English examples of the ditransitive construction. That construction can be divided into two patterns. The first indicates that one object noun is changing possession from the subject to the second object noun.

(5) John sent his landlord the check.

The second indicates that there is a predication between the two object nouns.

(6a) I found the guard sleeping
(6b) The guard is sleeping, and I found him.

(7a) We painted the room red.
(7b) The room is red because we painted it.

ἡγέομαι follows this second pattern. It takes two accusative nouns, and indicates that there is a predication between them. Phillipians provides several examples of this pattern and ἡγέομαι.

Phil 3:7
[Ἀλλὰ] ἅτινα ἦν μοι κέρδη, ταῦτα ἥγημαι διὰ τὸν Χριστὸν ζημίαν.
Yet, whatever gains I had, I consider these things loss because of Christ.

Here there is a a predication between ταῦτα and ζημίαν. “I consider these things to be loss.”

What we have so far in James 1:2 is, “Consider ____ complete joy.” There is only one noun phrase mentioned. We learned from Bergen that a reader will fill in these sorts of gaps with something that s/he thinks fits this space while reading. We don’t wait until we have all the information to put things together. We build with what we have and fill in the gaps with what we expect to fit until we come to that information.

No one knows what the first readers of James would fill in here, but if I was filling in the gap, I may think something like, “consider waffles complete joy”, “consider knowing Jesus complete joy”, “consider the love of God complete joy”, and so on. The presence of πᾶσαν χαρὰν constrains the reader to consider something, well, joyful! The reader will naturally fill in this space with something that they consider joyful. This builds an expectation for what is coming, and when it comes the reader is going to be befuddled.

But the reader has to wait to fill in the missing piece. James doesn’t fill in the gap immediately. What comes next in this string of words is a phrase addressing the audience directly, ἀδελφοί μου. Because this is the next set of words in the sentence, the reader has no choice but to continue to guess at what s/he is to consider total joy. The vocative phrase adds nothing new to the sentence, it only serves to delay the reader.

What fills the space in James 1:2 is the subordinate clause ὅταν πειρασμοῖς περιπέσητε ποικίλοις (whenever you  meet trials of various kinds). Wow! I wasn’t expecting that. And I’m willing to bet that the first readers weren’t expecting that either. Who would? To consider all kinds of trials to be total joy is counter intuitive. Trials and joy don’t go together…usually. But that is what James wants the readers to think. And as if the thought itself wasn’t dramatic enough, he uses the readers own expectations of what is joyful to add more effect to the command.

Had you ever thought about the limitations of language uptake and the way it affected meaning? It is something that I find fascinating, and hopefully I have demonstrated that it is useful in the study of scripture as well. I would love to go through other passages where this sort of devise is being used. If you have come across one please let me know.

Meet Jimmy…

Well, I finally convinced my husband Jimmy (@fakejimmy) to be a guest contributor here at Cataclysmic!  While he won’t be one of the regular Cataclysmic bloggers, he will be posting every now and then on his favorite subjects: linguistics and Biblical Greek.  Below is a short bio to help you get to know Jimmy:

jimJimmy Parks is a graduate of Houston Baptist University (MA in Biblical Languages) and will be pursing a PhD in the near future.  Jimmy currently works at a Maternal-Fetal Medicine office where he spends his lunch breaks reading Septuagintal Greek.  He also works as a student grader and occasionally substitutes for Greek, Hebrew, and Linguistics classes at HBU and SWBTS. During the summer he enjoys teaching Greek grammar classes at a local prison.  He is a deacon at First Colony Christian Church (Sugar Land, TX).  Jimmy is married to Jessica and they have two dogs – Charlie and Parker.

Jimmy is interested in Biblical Languages and Linguistics.  He loves reading books about language and the brain and is especially interested in how humans process language.

His first post will be up later this week so stay tuned!

In Which I Disagree With Mark Driscoll (and it has little to do with women)

Mark Driscoll of Mars Hill recently posted a blog over at The Resurgence on Why Mars Hill uses the ESV Bible. He starts with 6 theological reasons behind choosing the ESV (or why the ESV is the King of all Bibles).

I have my own qualms with the ESV (one of which he mentions, although he sees it as a plus) but I do own an ESV bible and use it at times when I need to read an English text. I consider it to be one among many readable translations. However, I found Driscoll’s six theological reasons for choosing the ESV highly problematic from a linguistic perspective.

What Driscoll puts forward as obvious truths I see as common misconceptions concerning Bible translation, interpretation, and hermeneutics that I hear quite often in the church.  The original post is rather long so I won’t quote it all here, but I did want to take a look at Driscoll’s third point and his reasoning behind it.

“The ESV upholds the truth that words carry meaning

Some scholars will argue that thought-for-thought and paraphrase translations do not change the meaning of Scripture, just the words of Scripture in an effort to clarify the meaning of Scripture. But this reasoning is misguided because meaning is carried in words. So when we change the words of Scripture, we are changing the meaning of Scripture.

There are a couple of problems with Driscoll’s reasoning here.  For one, linguists and other language scholars have argued the very opposite — that meaning is not carried in words.  I learned early on in my language studies that “words don’t have meaning.”

Fellow HBU graduate Kris Lyle has a great post (go read it!) on this very subject – Do words have meanings?  Kris points out the importance of CONTEXT in determining meaning because words do not come “pre-packaged” with meanings.  Linguistic research, particularly in the areas of semantics and pragmatics, has corrected this false idea that “meaning is carried in words.”

Kris provides a great excerpt from an article by linguist Vyvyan Evans (Evans, V. “Lexical concepts, cognitive models and meaning-construction,” Cognitive Linguistics 17:4 [2006], p. 492):

“That is, the ‘meaning’ associated with a word in any given utterance appears to be, in part, a function of the particular linguistic context in which it is embedded. Put another way, word ‘meaning’ is protean, its semantic contribution sensitive to and dependent on the context which it, in part, gives rise to (Croft 2000).

To illustrate consider the following ‘meanings’ of fast.

(1)   a. That parked BMW is a fast car.

b. That car is travelling fast.

c. That doddery old man is a fast driver.

d. That’s the fast lane (of the motorway).

In each of these examples the semantic contribution of fast, what I will later refer to as its informational characterisation, is somewhat different. In (1a) fast has to do with the potential for rapid locomotion. In (1b) it has to do with rapid locomotion. In (1c) it relates to ‘caused’ motion beyond an established norm: a speed limit. And in (1d) fast concerns a venue for rapid locomotion. Examples such as these show that the view of open class words, as possessing fixed meanings, is untenable on closer scrutiny. The precise semantic contribution of any word is a function of the utterance context in which it is embedded, and, moreover, the sorts of (conceptual) knowledge these lexical entities provide access to, as I shall argue in detail. In other words, words don’t have ‘meanings’ in and of themselves. Rather meaning is a function of the utterance in which a word is embedded, and the complex processes of lexical concept integration, an issue which is developed below.

Second, Driscoll argues that thought-for-thought translation changes the words and meaning of scripture by adding additional interpretation and commentary whereas a word-for-word translation does not.  Earlier in his post, Driscolls writes:

“A word-for-word translation (like the ESV) best enables this to occur by seeking, as much as possible, not to insert interpretive commentary into the translated text of Scripture.”

The problem here is that there simply isn’t an exact one-for-one correspondence between languages.  What I mean is, there isn’t one word in Greek that is perfectly and always represented by another word in English.  It is inaccurate to say that δικαιοσυνη means ‘righteousness’.  What we can say is that ‘righteousness’ is one of the English words that we can use in some instances to represent the Greek word δικαιοσυνη (depending on the context!)  Choosing a word from one language to represent a word from another language always involves a choice.  The truth is all translation involves interpretation, whether you employ a formal equivalency method (which Driscoll assumes is almost completely devoid of interpretation or commentary) or a dynamic equivalency method.  Formal equivalency translation seeks to represent as closely as possible the grammar and syntax of the original Greek or Hebrew while dynamic equivalency is concerned with representing the meaning or intention behind the original Greek or Hebrew as closely as possible.  (It’s interesting to see that both methods were employed by translators of the LXX throughout different books of the Hebrew Bible!).

In many conservative circles formal equivalency is often seen as ‘pure’ and ‘more accurate’ while dynamic equivalency is seen as ‘muddied’ and ‘less accurate’.  The truth is both methods have their strengths and weaknesses and both methods involve interpretation.

I imagine Driscoll’s post could generate a lot of discussion on Bible translation and the nature of language and it’s unfortunate that the blog does not allow comments.  What about you?  Did you find Driscoll’s reasons convincing?  Why or why not?  Is a word-for-word translation better than a thought-for-thought translation?

Runge on Porter’s Use of Contrastive Substitution

Steve Runge has posted a paper on his blog (to be published next year) critiquing Stanley Porter’s theoretical framework regarding the Greek verb. This is an important paper for readers of Greek and I highly suggest you check it out. Steve provides an explanation on his blog regarding the back story of why this particular paper is sorely needed:

“At the 2010 ETS meeting I presented an overview of some foundational errors in Stan Porter’s theoretical framework that significantly undermine the validity of his claims regarding the Greek verb. These issues initially came to light in research for my 2009 paper on the historical present.What I read left me with a knot in my stomach. Why? Well, Stan taught me second year Greek while I served as a TA for his first year Greek class at TWU. He was one of the folks who got me interested in linguistics in the first place, and he published my first article on Greek in one of his journals. I owe him a lot.

What was the big deal? The nature of the problems suggested a failure to adequately engage the linguistics literature. Significant counter arguments were ignored, as were warnings which should have led him to reach opposite conclusions about the presence of temporal reference in the Greek indicative tense-forms. One of the most significant pieces of evidence is the work of Stephen C. Wallace. I have posted his article, which is quoted at length in my critique. I would strongly encourage you to read it in its entirety. These problems were not just in his dissertation, but also in his recent writings on the prominence of the Greek tense-forms.”

Read more here: Porter’s Use of Contrastive Substitution