by Diego Román
Dr. Román looking very unscientific, dressed as a pencil.
I taught science for seven years to emerging bilinguals in Wisconsin and in California. My students struggled to understand the science textbooks we read, but their challenges went beyond comprehending the vocabulary or complex sentences used in those materials. When meaningful textbook-based learning happened, I noticed that my students were able to understand how conceptual information was presented by correctly interpreting how text segments (clauses and sentences) were logically connected to each other. Therefore, for my doctoral dissertation I decided to investigate the types of logical links between discourse segments in the context of middle school science textbooks. Let me explain what I mean by analyzing this sample text extracted from a science textbook:
1. Asteroids are left over from the formation of the solar system about 4.6 billion years ago. It is thought that they crashed into the inner planets during the early period of our solar system. Asteroids lack enough gravity to have an atmosphere. Consequently, their surfaces have many craters from impacts with other objects (Focus on Physical Science, Glencoe/McGraw-Hill, 2007, p. 489).
Interpreting paragraph 1, above, as a coherent passage requires students to assign meaning to this text as a whole. In other words, students must establish that the sentences in this text relate to each other in meaningful ways. In paragraph 1, one example of the types of the logical relationships established between sentences and clauses is signaled (or marked) overtly with the logical connective consequently. In fact, the meaning of that passage –that limited gravity results in a lack of atmosphere that renders asteroids vulnerable to frequent impacts that in turn results in numerous craters on the surface of asteroids– depends crucially on the meaning associated with the logical connective consequently, as evidenced by the alternative interpretation that arises when the connective is changed: “Asteroids lack enough gravity… because their surfaces have many craters.” In that case, it would be the cratered surface that is identified as the cause of the limited gravity.
The logical or discourse relation exemplified in paragraph 1 is one of inference—students must infer that the last sentence is the result of the explanation provided in the previous discourse segments. Inference is just one of the types of discourse relations that can be established between discourse segments. Although there are several ways to classify the logical links between textual segments, various linguists agree that these relationships can be categorized by four types of discourse relations or semantic categories: elaborative, comparison, inferential, and temporal—these categories are illustrated respectively by the logical connectives in addition, but, so, and then.
As shown in example 1 by consequently, discourse relations can be marked, or signaled, with logical connectives. When these discourse relations are marked, the main function of logical connectives is to guide readers in interpreting how segments of text are logically linked to each other. Yet, what is particularly challenging about understanding a text is that the relationships between discourse segments, quite often, are not signaled at all. Consider example 2:
2. Some plates include continents. The continents move with their plates. (Focus on Life Science, Glencoe/McGraw-Hill, 2007, p. 289).
In this second example, the discourse relation between the two sentences is implicit, in other words, no logical connective is there to guide students in interpreting how these two sentences connect. Therefore, in the text excerpt above students could think that the second sentence is providing a contrasting fact to the information presented in the first sentence as in ‘some plates include continents, however the continents move with the plates’; or providing an example such as ‘some plates include continents, for example, the continents move with the plates’; or correctly infer that the second sentence is a result of the fact described in the first sentence, ‘some plates include continents therefore the continents move with the plates’. This absence of logical connectives creates discourse comprehension challenges particularly to students who do not know much about the topic, struggling readers, and emerging bilinguals.
But, how often are discourse relations marked with logical connectives in science textbooks? In a recent study in which my colleagues and I analyzed a big corpus of science textbooks adopted in California, we discovered that only 2 out of 10 sentences in both subject areas contained logical connectives. Therefore, 80% of the time students need to infer the appropriate discourse relation by themselves. The implications of this finding are important. For one, our study may indicate that approaches that teach students to focus on lexical elements that appear in science texts might not be as effective as teaching students how language functions in science to accomplish specific goals (e.g., infer, compare, elaborate). And two, because science teachers possess the knowledge necessary to understand the implicit conceptual linkages in disciplinary texts, they are the ones best suited to make transparent to students how language works in their subject.
What can science teachers do about this? Although the task of analyzing how language is structured to present science knowledge could seem overwhelming, it does not have to be so. For instance, science teachers could focus on analyzing with their students two key sentences at a time, such as example 2 presented above, to discuss how language is used in their discipline to present essential conceptual information. Then, they can move to analyze longer paragraphs or sections of text. In this way, science teachers could develop meaningful opportunities for literacy and language development through the teaching of their discipline. In addition, teachers can adapt graphic organizers used in language arts classes to show how discourse is structured in science textbooks. Some examples of graphic organizers can be found here: http://www.readingrockets.org/article/how-teach-expository-text-structure-facilitate-reading-comprehension.
I hope this post illustrates that understanding of science texts requires students to be familiar with how language works in science to accomplish specific discourse goals and not only lexical items such as logical connectives that most of the time are implicit. As various authors have indicated, science texts require students to understand how language is used to construct arguments, generate relationships between claims, warrants and data, contrast and compare phenomena, recognize the nature and function of genres specific to the discipline, and identify the author’s purpose as a context for their response. This can be accomplished when science teachers model and socialize their students in the language demands of science texts by moving beyond discussions of vocabulary and syntactic complexity.
 I would like to thank Dr. Hannah Rohde at the University of Edinburgh in the United Kingdom and Ms. Stephanie Hironaka at Mills College in California for all of their work in these studies.
Share on Facebook