SLA: How to read deeply in Greek
Reading is a complex skill that involves the coordination of several different neurological processes simultaneously. In his article on reading in a second language, William Grabe defines reading as “a complex ability to extract, or build, meaning from a text” (Key issues: 8). Without the extraction of meaning, there can be no reading. This makes intuitive sense. As I write this, I am trying to convey thoughts through words by making use of symbols (letters) that represent spoken language. I am banking on you, the reader, to be able to decode the written symbols, so that you can understand my thoughts. For fluent readers, the process of decoding the letters, pronouncing the words, and extracting the meaning occurs in milliseconds for each word, and the extraction of meaning from written text is often as effortless as spoken communication. However, the story for disfluent readers is very different, whether they struggle reading in their native language or a second language. Readers who lack fluency “tend to read in a labored, disconnected fashion with a focus on decoding at the word level that makes comprehension of the text difficult, if not impossible” (Hudson, Lane, and Pullen 2005:702).
This description of “labored” reading is a common complaint from those learning the biblical languages, but it also hints at the solution to the problem. Disfluent readers are concentrating so hard on “decoding at the word level” that they cannot comprehend the text as a whole. In research on reading fluency, the term “decoding” is used of both fluent and disfluent readers, but the key difference is that fluent readers decode extremely quickly and effortlessly. It is not decoding itself, then, that it is the problem—it is how difficult it is for readers to decode. If too much of your limited mental capacity is devoted to decoding on the word level, your understanding at the text level will suffer because you are not concentrating as much on the text’s meaning—your attention is directed towards individual words. Our main point of this section is simple: the better you are at processes like decoding, the better you will be at reading.
Fluency, we will argue, is the key to making processes like decoding more efficient, which is the key to deep, meaningful reading. To quote Hudson, Land, and Pullen (2005:702-3) again: “The most compelling reason to focus instructional efforts on students becoming fluent readers is the strong correlation between reading fluency and reading comprehension.” After explaining the neurological underpinnings of deep reading, we will go into greater detail on how to harness fluency to rewire our brains for deep reading. In the last section, I contrast reading deeply with exegesis and suggest that the proper use of exegesis should not be primarily to uncover meaning but to describe meaning.
What it means neurologically to read deeply
Scholars have broken down reading into a number of different related, but distinct, processes. Although many have given much more fine-grained analyses, the following 8 steps provide what is happening in broad strokes when we read:
- Visual perception
- Orthographic identification
- Phonological representation
- Lexical retrieval
- Morphosyntactic knowledge
- Prosodic knowledge
- Connected-text knowledge
- Comprehension
Let us take the following sentence from Marathi to demonstrate how this works (if you know Marathi, pretend like you do not for a minute):
(1) मी मासे खात आहे
Mī māsē khāta āhē
‘I am eating fish’
In order to read this sentence, you first need to look at it (1) and identify the letters (2). When your eyes become accustomed to distinguishing one letter from the next, you will need to associate each of the letters with a sound (3). Once you can do the first three steps, you can pronounce what you are seeing, but you have not yet extracted any meaning. For that, you must learn what each string of sounds means (4) as well as how those morphemes combine together to form complex meanings (5). Next, you will need to know the proper way to read the text in context—is it being said seriously or sarcastically, as a question or an exclamation (6)? There is then the issue of how this particular sentence relates to other nearby sentences (7); it could be providing a reason for why you do not want anymore food, or it could be a reason why you want a drink, or in a more fantastical text, it could be a secret sign to your partner that now is the time to draw your sword and run out the door and slay a dragon. Finally, this particular sentence might be related to an entire discourse or story (8); it may be a mundane description of what you are eating, or it could be a pivotal turning point in your dragon adventure.
Steps 1-5 are usually described as lower-level processes and 6-8 as higher-level processes. This does not mean that one set is more important than the other, but the higher-level processes are more concerned with text meaning, and crucially, they are steps that cannot be automatized. When a skill is automatized, it no longer requires our conscious attention, so it demands very little cognitive effort. Good readers have automatized the lower-level processes, so they can focus more of their cognitive effort on the higher-level processes that concern how the text should be understood. In order to read deeply for meaning, we want to devote all of our cognitive efforts to meaning, and this means automatizing steps 1-5.
Reading begins at the ortho-phonological level (steps 1-3). In fact, statements like “I can read it, but I don’t understand it” really mean that the “reader” can perform steps 1-3 but nothing else. As William Grabe says, this step is foundational: “The automatization of letter-sound relations is the foundation of all alphabetic reading” (###Key: 9).[1] The automatization of this level gives you the ability to “read” something in your native language and be thinking about something entirely different. You have automatized the ortho-phonological level to the point where you no longer need to concentrate on pronouncing the words you are reading. The same should be true of your mastery of the Greek alphabet: you should be able to think about what you ate for breakfast this morning as you pronounce Greek sounds represented by Greek letters. When the ortho-phonological steps can be done without conscious attention, you will know that you have mastered these steps.
Steps 4-5 represent morpho-semantic processes that connect sounds to meaningful units. Again, Grabe notes how important these processes are: “fluent readers have very large and automatic recognition-vocabulary knowledge and…vocabulary knowledge is highly correlated with reading ability” (###Key:9). Knowing the words in this sense is to know the meaning of the words (step 4) and what kinds of morphemes and syntactic environments co-occur with the meaning (step 5). To automatize both of these steps, you need to be able to immediately think of the meaning of a word like ἅπτομαι and automatically recognize that the genitive that follows it is an argument of the verb. Biblingo utilizes spaced repetition technology and fluency practice to automatize step 4, and we teach the verb with its arguments and encourage extensive reading to automatize step 5.
The higher-level processes represented in 6-8 are not automatized; indeed, they cannot and should not be. There is no way of knowing whether the sentence I am eating fish is a secret sign or a mundane description of your dinner without knowing what is happening in the rest of the discourse. Whereas we do not want to be thinking about what sound the letter φ makes as we read or even the meaning of the word φύσις, we do want to be thinking about the relationship between the current sentence we are reading and the surrounding text. To read deeply is to devote all of one’s cognitive capacities to determining the meaning of the text. Indeed, the brains of fluent readers operate differently than disfluent readers in this respect, and it is all based around how much time we spend on lower-level processes. Wolf (2007:143), quoted in the preface, is worth quoting in full once again:
“The fluent, comprehending reader’s brain is on the threshold of attaining the single most essential gift of the evolved reading brain: time. With its decoding processes almost automatic, the young fluent brain learns to integrate more metaphorical, inferential, analogical, affective background and experiential knowledge with every newly won millisecond. For the first time in reading development, the brain becomes fast enough to think and feel differently. This gift of time is the physiological basis for our capacity to think ‘endless thoughts most wonderful.’ Nothing is more important in the act of reading.”
– Maryanne Wolf
From this discussion, we know what we need to do in order to become deep readers of Biblical Greek—we need to automatize lower-level reading processes in order to devote more time and attention to higher-level processes. The next big question is: how do we give our brains this all-important gift of time?
Harnessing fluency to develop deep reading
William Grabe defines reading fluency as “the ability to read rapidly with ease and accuracy and to read with appropriate prosodic word stress and phrasing while understanding the text” (Reading 2nd lang: 404). Without the ability to read “rapidly”, there can be no fluent reading, since we are not giving our brains enough time to devote to the higher-level processes. We need to get faster, but how do we do this?
Going back to Paul Nation’s Four Strands of language learning, the fluency strand must involve material that is easy where all the vocabulary and grammatical constructions are known. Fluency development is not the time to learn new things; it is the time to get faster at what you know. One of the most well-studied fluency development drills is timed repeated readings. As far back as 1979, S. Jay Samuels proposed the method of repeated reading to increase comprehension and reading speed. The idea was that students would take a short passage and try to read it at a certain speed measured in words per minute (WPM). They would read it repeatedly until they reached the desired WPM, and then they would move on to the next short text. What Samuels (1979) found was that students would start off at a faster speed with each new text they started the drill with, and the number of times it took to get to the desired WPM would also decrease. In other words, timed repeated reading of a passage increased the participants’ baseline reading speed. This is, of course, exactly what we are after.
Repeated reading is great if you can already read, but many are still stuck at the alphabet. As Grabe said, “The automatization of letter-sound relations is the foundation of all alphabetic reading,” so this lowest-level process must be mastered before you can begin to do repeated reading for fluency development. However, the method of repeated reading and Nation’s requirements for fluency drills can also be applied to building letter-sound automaticity. You need to practice with known letters and focus on going as quickly as possible until you reach a certain desired speed with each letter. Again, you want to get to the place where you do not have to consciously think about the sound when you see it. You say it automatically with no hesitation.
Going one step up in the reading process, word-recognition should be practiced in the same way. This can be done with flashcards, but it should be done with words that you should not struggle with. The point is to go through the words very quickly. The faster you can recognize a word, the faster you will be able to read that word in a text. A crucial part of increasing fluency with vocabulary is that it must be done with every new word that is learned. As you increase your ability to read more complex texts through the acquisition of new words, you must also increase in speed with those new words. If fluency is not deliberately practiced, it is easy to remain mediocre at a large amount of words. In order to read fluently, you not only need a large vocabulary, but a large “automatic recognition-vocabulary knowledge.”
There are, then, three stages you should focus on to increase speed: letter-sound correspondence, word-meaning recognition, and reading full texts. Depending on where you are in your language-learning journey, you may need to focus more on one of these processes over the other. In general, letter-sound correspondence must be mastered before the other two, and more time should be devoted to word-meaning recognition until you know enough words to read a variety of texts. By increasing your speed and proficiency at each stage of the reading process, you will dramatically increase your overall reading speed, and you will give your brain the time it needs to comprehend and feel a text as it was meant to be comprehended and felt.
Reading deeply as an alternative to exegesis
We have outlined a path to deep reading through fluency development. This discussion of deep reading begs the question: how does deep reading relate to exegesis? In the biblical studies world, there are two different definitions of exegesis that are regularly given, as exemplified by Douglas Stuart (AYBD) and Donald Hagner (NT exegesis and research):
Stuart: The process of careful, analytical study of biblical passages undertaken in order to produce useful interpretations of those passages…The goal of exegesis is to know neither less nor more than the information actually contained in the passage.
Hagner: The goal of exegesis is to articulate the meaning of a passage as it was intended by the original author.
Although similar, these definitions reflect different relationships between the meaning of the passage and the exegete. According to Stuart, exegesis is a process that “produces” interpretations, i.e. we use exegesis to determine what a text means. In contrast, Hagner’s definition presupposes that the exegete already knows the meaning of the passage and is trying to “articulate” that meaning. Because exegesis is itself a scholarly construct created by biblical scholars, there is no one correct definition of it, but I do think one is more useful. In general, what we call exegesis should, in my view, be about articulating meaning and not determining meaning. You should determine the meaning of a passage primarily through deep reading.
There are two primary issues with using close analysis to determine meaning. First, all humans are naturally much better at using language than analyzing language. To illustrate, what would you say the meaning of the -s morpheme is on the end of cats? If you are like everyone else I have asked this question to, you probably immediately thought more than one. However, this leaves unexplained the use of cats in Zero cats are eating. All native English speakers intuitively agree in their usage of the “plural” morpheme with the number zero, but they almost all also agree that –s contributes the idea of being more than one if asked explicitly about the meaning. In other words, their language intuitions disagree with their language analysis, and this is despite the fact that they are native speakers of the language who have accurate intuitions about the full range of meaning. When it comes to a second language, analysis is even more difficult because we do not always know the full range of meaning. However, when a non-native speaker of English reads the sentence Zero cats are eating, they can determine the meaning as long as they know how the word zero is used. In other words, non-native speakers can build comparable (though not identical) intuitions to native speakers, and they can certainly become quite proficient at understanding native speaker output (such as in written texts). Being able to analyze a language and accounting for its full range of uses is much more difficult.
The second issue is with how texts are analyzed when doing traditional exegesis. Commentaries and biblical scholars will work through a text word-by-word or even morpheme-by-morpheme, and the analysis becomes very atomistic. In this kind of “exegetical process,” the forest is often missed for the trees. Romans 10:4 is a prime (exegetically significant) example. Books have been written on (and named after) this verse, but it is widely misunderstood. The verse reads τέλος γὰρ νόμου Χριστὸς εἰς δικαιοσύνην παντὶ τῷ πιστεύοντι ‘For, to everyone who believes, Christ is the end of the Torah for righteousness.’ In this translation, I have fronted the phrase παντὶ τῷ πιστεύοντι to make the meaning clearer. When read this way, the verse says nothing about Christ being the end (τέλος) of the Torah, and the meaning of the word τέλος, which has been the subject of a tremendous amount of debate, is not nearly as critical. Commentators and biblical scholars often read the verse as if all it said was τέλος γὰρ νόμου Χριστὸς, and they go on to talk about salvation-history and in what sense the Christ would be the end of the Torah. This, however, is not what the verse says, and it is a fundamentally different assertion. The verse is not about salvation-history; it is about the perspective of believers. They no longer attempt to find righteousness in the Torah when they have found it in Christ. Ideas like the Christ abrogating the Torah in some sense or the Torah culminating in the Messiah or the covenant climaxing in the Christ (or whatever variation of these you choose) might have some validity to them, but Rom 10:4 cannot be used to support them. In contrast to these modern scholars, Origen rightly states in his commentary that the Torah ending only applies to believers (##). I do not at all say this sarcastically or to denigrate modern scholars, but Origen kept reading. As a fluent reader of Greek, he read through the text instead of picking it apart word-by-word, and he, therefore, recognized that this verse was not a universal statement about the τέλος of the law. It is about how believers view the law. By focusing so much on the word τέλος, scholars have neglected other parts of the clause and have misinterpreted this pivotal phrase.
I am not suggesting we throw out exegesis. I am suggesting that deep reading (i.e. continuous reading and re-reading of passages at near native-speaker speed) should be the primary way we determine the meaning of a passage. Grammatical categories are helpful to explain the meanings we intuit by proper use of the language through accurate reading. For example, it is helpful to have a category of meaning for the dative παντὶ τῷ πιστεύοντι, namely the perspectival dative, after we read through the text and recognize that as its meaning (see ## for more on this use of the dative). It is also helpful to look more carefully at these kinds of datives by reading through passages with different examples to get a better sense of what these constructions are doing, but then we must go back and re-read the passage. Building out a system of grammatical categories and functions will then help us to articulate the meanings we are intuiting in good, deep reading. Sometimes careful grammatical analysis may also help us to decide between readings when multiple readings are debated and we need to make a case for one reading of the grammar over the other. However, for the vast majority of people, building strong Greek intuitions and reading deeply will be the best guide to determining the original meaning of the text.
[1] Even when you read silently, you are normally sub-vocalizing as you read. This means that you are “pronouncing” the words in your head, and in fact, even your muscles are firing that would be used to pronounce those words. They just aren’t firing hard enough to cause movement (###).
