Structural Assessment of Learning Object

This text is an adaptation of an excerpt from: Roth, W.-M. (2010). A social psychological reading multimodal scientific texts in online media. Reading Psychology, 31, 254–281. (You can access the original with your UVic access here: https://www-tandfonline- com.ezproxy.library.uvic.ca/doi/pdf/10.1080/02702710903256411.) I did this work because I wanted to know the resources made available by online materials that enable reading for understanding. The text provides some conceptual tools for the analysis of online media contents, all exemplified in an analysis of a BBC online science article (Figure 1, p. 2). The article provides some answers to the problem researched in Block 1 readings. This analysis teaches you a lot about design issues with respect to how readers / viewers approach learning objects. It teaches you what in reading we often do without even being aware of it.

A competent news reader is one who can: follow narratives across columns and pages of texts, tell apart such objects as news stories, news features, editorials, advice columns and advertisements, relate what is said today to what was said yesterday, and so on.

Over the past decade, some science educators have come to realize that reading in science has received a short-shrifted treatment. If acknowledged at all, the main focus of instruction lies on vocabulary development. Even though textbooks are the predominant resources in science instruction, reading in science has not been a major concern for teachers and researchers alike. Yet many people read science-related articles in their daily newspaper, science-related fiction, or in online news features. Thus, topics such as nanotechnology not only lead the general public to read more books in science technology but also such texts incite many members of society to become new readers of science. News coverage of certain scientific problems such as climate change has increased exponentially over the past decade or two, which, given the demand-driven market forces that the media are subject to, reflexively means that the public consumption of science-related features has increased. This is an interesting trend in the face of evidence that science instruction continues to turn students off. If there are new science adepts because of publicly available science texts, then even novice science readers already enact (sets of) skills and practices that allow them to understand science-related texts with and for understanding. What are such skills and practices that allow reading to organize itself in a novel domain, here science, often said to employ language with particulars that make reading difficult? If members of the public generally are able to be uninstructed new readers of science and science-related texts, then the prerequisite practices and background understanding have to be available in society broadly. By participating in society, members concretely realize intersubjectivity, common knowledge, and cultural practices in their daily actions in recognizable (correct) ways precisely because any higher psychological function was external and therefore available to anthropological study. Any particular reading of online materials therefore merely realizes possibilities already available in culture. These same cultural competencies allow readers to read this text, though it is a different genre from what they are familiar with and used to. Any set of particular reading does so concretely, thereby revealing cultural possibilities generally. By enacting within- individual readings, we therefore get at not only the range of readings but more importantly the conditions for any one particular reading more generally. As every morning, I begin the day of February 13, 2007 with a quick look at the home page of BBC News (news.bbc.co.uk). There are headlines with one-sentence texts and images on the top half of the page, a line of images, titles, and text under the rubric of “Features, Views, Analysis” in the center part, and regions or topical rubrics with one or two bulleted items in the bottom part. All headlines and bulleted items are in blue, the color that links to other pages take on my screen. I scan the titles and bullets, and click on some, but not on others often within the same category. This morning I click a hyperlink that reads “Ancient chimps ‘used stone tools’” and is listed under the category “Science.” It takes me to a page that bears as title the same phrase as the hyperlink that has taken me there. In addition to the title, the page contains one boldfaced 1-sentence paragraph, 10 Roman printed 1-sentence paragraphs, one subtitle, and two captioned photographs (Figure 1). But before I actually read the text, I begin to reflect asking questions such as, “What was it about this title that has made me click the link to read the associated article or at least check out whether I want to read the article in its entirety? Why did I click on the link and read the article? Why did I go to the BBC website scanning the titles, texts, and images and select this one over many other ones? What is the purpose of this reading? What effects do this reading have on my

private and professional lives? And equally importantly, what are the practices that allow me to make sense of the titles and the texts that make me return on a daily basis? Rather than actually read the article, I begin to reflect about possible answers to this question.

Figure 1. Offprint of a science article published online by BBC on February 13, 2007. (Permission granted by the BBC on November 7, 2007) You can access it here: http://news.bbc.co.uk/2/hi/science/nature/6356773.stm

The hyperlink constitutes an invitation; and like all invitations there are those we follow and others that we do not. My cultural competence allows me to recognize the invitation as invitation, and this recognition is realized and expressed in the clicking of the hyperlink. I am not likely to click a link containing the name “Britney,” especially when it is categorized under “Entertainment,” which is likely but another story about the stupidities of Britney Speers, which, from my perspective, ought not be making (the) news in the first place. I am not likely to click the link “Americans turn to online videos” (news item on January 10, 2008) but I may, depending on many other mediating elements, click the link “New study says 151,000 Iraqi dead” (January 10, 2008) especially at a moment that the false pretenses under which the Iraq war was begun are salient. That is, it is not the reference to America, Americans, and American culture that determines whether I follow a link, but an interaction of a variety of circumstances, including the predicates linked to the subject of the topic. A hyperlink seeks to interest, articulate newsworthiness by announcing that something we need or ought to know is provided in the text to which it leads us if we click it. The hyperlink does so, like a headline, by announcing a continuation or conclusion to something we are already interested in or by announcing that something extraordinary has happened. But the interest cannot be in the link, as there are other hyperlinks equally designed by the same editors that I, as others, do not follow. Interest therefore has to be the result of a mutually presupposing text|reading constellation; but results presuppose work, however much it is or has been made invisible. Reading, even if it only concerns a hyperlink on a website, does work. This work is so ordinary that we normally do not reflect upon what we do and how we do it. After all, I did follow the hyperlink without stopping to reflect, but I did not follow the hyperlink mechanically, as there were many other links that I did not follow. This work of reading inherently has a public (social) dimension, which is the source of all individual cognition. The social psychological (anthropological) project aims at understanding the often-invisible work of cognition, here reading, by rendering the familiar strange. Even attitudes and interests need to be approached from a social psychological, sociological, and cognitive anthropological perspective rather than from within the individual and by means of a framework that considers cognition as something private. The purpose of this text is to present an approach to the social psychology of reading grounded in cultural-historical activity theory and one particular branch of sociology, ethnomethodology. To achieve this, I bring together ethnographic descriptions, cultural artifacts, and concrete descriptive analyses to exhibit more so than to explain a social psychology of reading. Here, I am not concerned with my personal, singular reading or with a collection of specific readings (and correlation to knowledge, interests) some population sample may provide. I am not interested in whether any part of the BBC website I studied can be interpreted in this or that way; nor am I interested in the different ways in which various readers interpret the text—this work I leave to phenomenography. Furthermore, I am not interested in a social constructivist analysis that focuses on the media’s representation of science, ideology, and the politics of science. Instead, what I am interested in is the more general and generalizable question of how this or that reading is enabled, the resources and conditions that afford (enable) the different types of text|reading pairs to occur in the first place. That is, in this article I focus on the cultural practices of reading and the possibilities that they constitute; and in my reading, these general possibilities are also concretely realized. A coherent and comprehensive social psychology of reading is a very large undertaking—necessitating, for example, hitherto untried marriages of the technical and the sociological literature.

On Finding Newsworthy Science Content in Hyperlinks/Headlines In this section, I analyze the lived work of reading that each and every reader accomplishes following online media hyperlinks such as “Ancient chimps ‘used stone tools’.” Reading the news is not an invitation to engage in a post-modernist game of interpretation where any reading (interpretation) is as good a reading as any other. I do not have the license to invent any form of possible meanings, at least not if I want to report “the news” to other individuals and groups. News headlines are not “open texts” (e.g., like James Joyce’s Finnegans Wake) or poems that await a range of interpretations. Reading news for the

newsworthy events that they have been written to communicate requires specific readings rather than readings where anything goes. Science and scientific literacy is deployed beginning with the reading of a hyperlink that takes us from the main page to the page with the specific article announced in and by the hyperlink. Because some hyperlink hooks novice science readers, understanding the way such links function is an important step to understanding reading literacy online materials. Headlines, which in print media appear directly above the article, are used in online media also on the main page without or with limited amount of text and images. That is, these hyperlinks are to be read as announcing the contents of the news and to draw readers to the stories, which are accessed by clicking (following) the hyperlink. Because of the limited amount of space available the hyperlink (headline) cannot say everything that the associated story says; in fact, the hyperlink does not have to say everything because the story itself is saying it. But the hyperlink (headline) is (needs to be) such that it draws the potentially interested reader to the story itself, which is accessible only by “following the link,” enacted by means of a movement of mouse-cursor and clicking the hyperlink once the cursor hovers over it. In fact, for novice readers who eventually become interested in some science topic, reading the hyperlink text has to configure the process of reading itself for interest to emerge. On Seeing that Chimps Living a Long Time Ago already Used Stone Tools The hyperlink reads, “Ancient chimps ‘used stone tools’.” One can read the this hyperlink (headline) to be about chimps that lived a long time ago, perhaps at the same time or before humans (homo sapiens) or one of its predecessors; these chimps apparently have been documented to use tools either by picking up suitable natural materials or by modifying stone splinters. How does one arrive at such a reading, which, as a possibility, is figured both in culture (after all, the news editors figured the text) and in the concrete realization of its possibilities in this and similar readings? What in this hyperlink (headline) is it that it is, and announces something, newsworthy? Answers to the questions are important for understanding reading, as I do not inherently click on every hyperlink featuring the word “chimpanzee,” even though I may be interested in chimpanzees as a particular context for thinking about the emergence of culture and the specifics of human culture that makes it different from cultural rudiments observed in animals. I do not attempt to find articles, even the latest ones, on chimps and chimp culture, though when I see one then I will download or immediately read it. These are ancient chimps, not the ones that have been used in primate research since the work of the gestalt psychologist Wolfgang Kohler had done on the mentality of this primate or the ones that Jane Goodall brought to the attention of millions of scientists and just plain folks alike. These are ancient chimps that predate science and record keeping so that the primates themselves must have left records in and through their practices that allow researchers today to make inferences about the chimps living a long time ago. In the hyperlink, these ancient chimps are said to have used stone tools. They are said to have used the tools rather than definitely used the tools: this reading outcome is the result of the quotation marks that we read as reporting (quoting) the speech (claims) of someone else. Tool use is part of the news, but tool use itself is less than certain, made as such by the use of inverted commas that constitute a form of modalization common among scientists when statements about some phenomenon do not go uncontested and are still some way away from being taken as fact in the community. Chimpanzees today, those that Jane Goodall and others have been writing about, frequently use plant-based tools, such as leaves or parts of branches stripped of leaves to pull termites and ants from their habitations (ant hills, termite mounds). “Stone” can be read in contrast to these wood-based tools, a reading consistent with predominant anthropological discoveries of tools from non-decaying inorganic materials—stone, metal, shards from backed clay. The reading of “ancient” as “a long time ago” therefore is supported by the material nature of the tools, which is from non-decaying inorganic rather than decomposable organic materials. Historically, “tool use” has been a predicate solely attributed to humans; in recent years, an increasing amount of evidence shows that not only primates but also birds (e.g., the New Caledonian crow) may use tools. The link and distinction between humans and animals based on tool use has been ascribed to

Benjamin Franklin, but other scholars have variously expressed this idea including Karl Marx: (Hu)mans are tool-using animals. In the present context, tool use stands out because it is linked not to animals (chimpanzees) generally but to ancient chimps specifically. We can read the predicate “tool-using animal” as establishing a commonality (category) and a distinction: humans are animals, but they distinguish themselves from the latter in that they use tools. Here, then, not only do chimps use tools, which many readers may already be familiar with, but also they did use tools and these tools were made of stone. Finding “The Discovery,” that is, the Newsworthy Item If reading has not found the newsworthy part by the time it gets to the first paragraph in normal (rather than boldfaced) type, it is provided with further resources in the first two words of the regular text. The term discovery literally means “taking away a cover,” which, as covers generally do, hide things from our gaze. A discovery reveals something not known before, and therefore constitutes something new and perhaps unexpected. As definite article, the “the” allows us to read “discovery” as something already known or announced. If the text linearly proceeds from left to right and top to bottom, whatever “discovery” denotes already is/should be familiar to the readers and has to be some common knowledge or has to be available in the text itself. The previous sentence—in the paragraph appearing in boldface type—constitutes a statement about chimpanzees having used stone tools a long time ago, which, in fact, is an elaboration of the title. The discovery has to be found in this statement, which, to the knowledgeable reader, constitutes a fact deserving the denotation discovery. The “The” therefore allows reading to organize itself such as to seek what it is that we already should know, or reorganizes any previous reading of a mere statement that now becomes a discovery after a fact. The “The” therefore also constitutes an integral part of a pedagogy that allows reading to recognize the discovery in what the previous statement describes. The use of the definite article in this way is prevalent throughout the remainder of this text (“The skills,” “The tools” [caption], “The excavated stones,” “the age of the tools”).

Cultural-Historical Resources Available to Reading What allows me and any other reader to produce a reading are available contextual resources, that is, categories, category collections, devices, and predicates that online media (here the BBC homepage) make available for the reader looking for “the news.” In particular, a social psychology of categories, category collections, devices, and predicates focuses on how members of society, thought of as lay and professional social analysts, use membership categories, membership categorization devices, and category predicates to accomplish naturally occurring, ordinary, everyday activities. Membership categories may be interactionally linked together to form families, collections, or classes referred to as “membership categorization devices.” In the following, I discuss the main cultural resources that allow reading to do its work. Categories, Category Collection Devices, and Heuristics In the present context, whether explicitly named or thematically present, “animals,” “humans,” and “primates” all fall into a membership categorization device, because they “go together.” This specification to the present context is important because of the contingent ways in which members of society produce and use membership category devices for the purposes at hand. We cannot therefore in a definitive manner speak of what a membership categorization device consists in without saying what it consists in this time. There are two heuristics for using membership categorization: adequate reference (sometimes also economy) and consistency. The first heuristic means that a single category reference suffices to characterize a person or thing. Thus, it suffices to categorize chimpanzees as mammals; an additional reference to their inclusion in the animal category would be read (heard) as superfluous. The heuristic holds that once a first category from a category device has been used, other categories from the same collection may be used to categorize further members of the population. Therefore, once the term “mammal” has been used, other terms from the same collection (“animal,” “human”) may be used.

The particular reading of “Ancient chimps ‘used stone tools’” is in part configured by the fact that the link appears under the heading of Science rather than under “Americas” (the researcher making the finding works at the University of Calgary”), “Africa” (the findings were made in Ivory Coast), “Technology” (tool use), or general news. Following a link that appears under the heading “Science” allows reading to configure itself to anticipate a particular genre of text rather than a poem, an open literary text, and so on. It also configures reading to produce understanding of content in science-specific rather than in other ways, for example, as science fiction, in which case the article might appear under the heading “Entertainment” (as it would be if the article reported the production of a sequel to or remake of the Planet of the Apes). The boldfaced red-printed terms “Science,” “Technology,” “Entertainment,” “Business,” “Health,” and so on that appear on the BBC homepage constitute category collections, which, as such, bring together a great variety of articles and hyperlinks that may not go together in other instances. A first maxim states that if two categories are used to categorize two members of some population, and those categories can be read as categories from the same collection, then should be read that way. Thus, and relevant to the present instance, humans, tools, and tool-using animals may be brought together into a (named or unnamed) category collection. Because the hyperlink “Ancient chimps ‘use stone tools’” appears under the heading Science, we already and without having to reflect anticipate and expect the newsworthy item to be a scientific one also of interest to a general audience—not all scientific discoveries are reported in the general news media but rather remain relegated to the domain- specific journals. “Science” therefore constitutes part of the background that configures reading to read the headline in a more constrained way than if the “category” title had not been present. “Science” is the background against which “ancient chimp ‘uses stone tools’” is read and against which it becomes the figure that it is, that is, takes on the salience that it does. Science constitutes a particular category collection device, which organizes the materials at hand to allow some readings more so (rather) than others to emerge from the text|reading pair. Standardized relational pairs make for category collections including, for example, husband-wife, teacher-student, or perpetrator-victim. The standardized relational pair allows reading to invoke one member of the pair once the other is present. The members of pairs are related to one another; the pair in the present case is a category of inclusion (animal) and a category of exclusion (human versus animal). The contrast human-animal constitutes a standardized relational pair. Another membership applies in this particular situation is that of “online science news” (a collective category that also includes the other websites I regularly peruse: CBC, Le Monde, and Die Zeit) and “science news reader” (here I, as concretely realizing possibilities of reading and understanding). These two are parties to an online news feature reading, an example of a “standardized relational pair of categories.” The knowledge and orientation associated with this pair may be considered to be procedural, embedded in a practice of “looking for the news,” that is, reading in a way that allows me to discover the newsworthy story. Science news producers and newsreaders are oriented toward the properties of phenomena and findings that make them newsworthy rather than something else, literary texts, poetry, and so forth. Part of the newsreader’s activity is the constitution of the other party, here the “BBC” or its journalist, as an actor and participant in the transactions. What the “BBC” reports about the chimpanzees is a reportable and accountable matter, for example, when I talk about the feature to my colleague in New York using iCHAT or to my wife over dinner after she has returned from work: “Did you know …?” The category pair “online science news” and “science news reader” centers on the collection of tasks and cultural practices that make up “reading what the BBC says about and constitutes as reportable and newsworthy in the world.” Predicates: Grammar of Actions for Identifying Membership The predicate “used stone tools” normally is associated with attributed to (early) human beings, especially those that lived during what now is referred to as the Stone Age. That is, the action of using (stone) tools, the predicate in the link “Ancient chimps ‘used stone tools’,” is bound to the category of being human. Many actions are bound in this way to specific categories, such as flying (birds), barking (dogs), painting (humans), or growing (living beings). Thus, the action of using tools had long be hailed to be a specifically human activity—until evidence, such as the one referred to in the online text analyzed

here dating to the 19th century, showed that animals too used tools. A category is invoked when a category-bound activity is asserted leading to the binding of category and activity. This binding therefore allows and affords categorization when the action is named or seen. A description of this work of getting to the category once a category-bound activity is seen has been captured in a “viewer’s maxim” that ties together the seeing of a category-bound activity and the member of a category to which the category is bound. That is, the doer is seen as a particular kind of individual, namely the kind (category) of individual who does engage in the form of actions/activities observed. But because tool use is relegated to human beings, its association with an animal (primate) elevates the latter from other members of the animal kingdom. That is, tool-use elevates the user from among mere animals, making it, depending on context, a special animal or something human-like, for example, a hominid, humanoid, or anthropoid. That is, the category-bound activity allows reading to co-select or partition the categories, into those who are able or normally engage in it and those who are not. An increasing number of species formerly classified as animals in contrast to human have been associated with “intelligence” and “culture,” which used to be typical and quintessential human characteristics. Thus, there are scientists studying communication among orcas and other whale species, cultural and society-specific practices that are handed from one primate (chimpanzees, orangutans) generation to the next, studies of trading (food for favors) among bird species, and tool-fashioning, tool-use, and handing-down tools among (New Caledonian) crows. Therefore, reading that ancient chimps have used stone tools now has the possible meaning that tool use never distinguished animals from humans but is a skill more ancient than the human species (homo sapiens).

Reading a Multimodal Science Text In the foregoing sections, I describe the resources available to the work that accomplishes finding the news in the hyperlink (title). This work already mobilizes many of the resources required in reading the text itself. But the text provides further structures that reading mobilizes both to configure itself and to find the particular content the text was written to convey. New resources become available as soon as we get to the article (Figure 1): the first three paragraphs are of one-column width, sharing the space with a black-and-white photograph of entities that can be recognized as stones, one of which easily may go for a spear tip. The display provides a structured terrain that provides resources to reading to organize itself, structures that a physical geography of reading allows us to uncover. I articulate the additional resources in the following exemplary reading of the remainder of the article. A Physical Geography The display is not the same throughout but physically differs in its different regions. Such differences, constitutive of the regions, are resources that allow cultural practices of reading to separate out and relate different parts of text. Generally the text is black, but below the images it is printed in grey. The text sizes differ, being largest in the first line (which we recognize as title both their position—English, as other European languages is written from top left toward bottom right, left column before right column). A second type of text is smaller, but still larger than majority of text, differing from the latter in that it is printed in boldface type similar to the title immediately preceding it. This text, therefore, physically constitutes a transition, sharing physical characteristics with the text preceding and, in length, with the text succeeding it. It is a piece of text and an elaboration of the title: a non-identical repetition (elaboration) of the title and a subtitle. Further below, there is another text of the same nature, that is, larger than the majority of the text and in boldface type (“Nut crunch”); it differs from the previous text of the same physical characteristic in that it only contains a compound noun (in other instances, a noun and an adjective) rather than consisting of a full sentence. The text beneath figures is smaller still than the main text and always appears in the column not exceeding the width of the image. Competent reading sees it as different and associated with the image rather than constituting part of the main text. That reading sees caption and the remainder of the text as different may be unremarkable and overstating some point. Yet we may gain a new appreciation of this relation in light of the fact that copy functions in computing environments—e.g., in (scanned) PDF materials—where the different columns remain unrecognized by the optical character recognition

program. Again, therefore, the attribution of the grey text to the image—that is, the relation between the two—is the result of the work accomplished by the text|reading pair rather than being exclusively to be found in the text or in the reading. There are other physical structures as well, which provide additional resources for the reading to configure itself. Among these we find what are known as commas, periods, inverted commas (title), quotation marks (third paragraph from the bottom, and inverted commas. These can be made to work together with the text, such as when the inverted commas around “used stone tools” can be read as decreasing the degree of factuality of the statement they enclose much in the same way as the clauses referring to their produced nature of (e.g., “say researchers” [caption 1], “researchers say” [¶8], “write the researchers” [¶9], “the authors say” [¶3]). Furthermore, not all letters are of the same type: some are recognized as capital letters in contradistinction to small letters. Etymologically, “capital” means standing at the head; and letters at the beginning of a paragraph or chapter in certain literary books are not only in capital type but also decorated, many times the size of the remaining print. Capital letters stand at the head of words (names) and sentences in much the same way that subtitles and titles stand at the head of text sections and entire texts. Reading knows a beginning to occur, a new sentence or paragraph, and with it, the possibility of a new topic. Capital letters also lead us to read some words as names, even though they may also exist as mere words (nouns, adjectives), such as, for example, the words “Proceedings” (third to last paragraph, Figure 1) or “Stone Age” (second to last paragraph). Here, the capital letter constitutes a resource that allows reading to read a name rather than a category word. This is the case even when there is a word that readers may have never encountered before: Even when reading “Noulo” for the first time, capitalization allows it to be recognized as a name (especially together with the definite article “the” discussed below) rather than as an adjective modifying “site” (last paragraph before “Nut crunch” subtitle). On Bridging The different characteristics not only mark out physical terrain, but they also mark out conceptual terrain. In the course of its unfolding local history, reading organizes itself to establish relations, bridges, between the different parts so that from the organized whole emerges the sense of one narrative. The most basic technique for establishing a bridge between multiple pieces of the same type (within main text, titles, captions) and different type of text (across title, caption, main text) is the preservation of a category or category membership device across the spatial and temporal gap in the text|reading pair. Thus, action verbs (e.g., “to crack nuts” [¶1], “tool use” [title, ¶2]) and skills (“The skill” [¶3]) are from the same collection, which allows us to read the third paragraph as making reference to the topic of the title and the first two paragraphs. The bridge between text and image, however, is more complex because there is a translation between two very different domains of reading is involved (see below). Yet even the replacement of a word by a synonym constitutes a translation, which, as all translation, relates two things that are non-identical and therefore not replicas of one another. Thus, reading has to exact work to make “Julio Mercader and colleagues” (¶6), “the authors” (¶3) the agents of the report in “the Proceedings of the National Academy of Sciences journal” (¶4), and “the researchers” (¶9, caption 1) all pointing to the same group of individuals. To produce a sense of oneness, the text|reading pair has to provide resources (text) and linkages (reading) that relate different parts of the multimodal display (Figure 2). These bridges and links are not in the display: otherwise they would not have to be made. But they are not in the making (reading) alone, because then the reason for making them could be found in the reading practice itself and it would not require a or the text. It is in the dialectic of the text|reading pair that the bridging links come to emerge as the contingent product of this reading this text.

Figure 2. Category repetition and category complexes constitute resources for producing connections between different (textual, non-textual) parts of the online display. Image|Caption Photographs provide a textured surface that provides resources for reading. On their own, photographs share a lot with poetry in that they lend themselves to very different even oppositive and contradictory readings. This clearly was exhibited in an analysis of the same images environmentalists used to claim that some area is environmentally degraded (polluted) and that it has already undergone a lot of restoration work. The texts in the captions make the difference for seeing a photograph in one way rather than in another—or, equivalently, for reading to find entities in the photograph that are expressions of specific concepts. The image and the caption then stand in a dialectical relationship, where the caption states what can be seen but the photograph confirms the appropriateness of statement in the caption. In this case, the photograph motivates the reading of the description in the caption and the caption motivates the reading of the image. However, the relationship between image and caption is not an easy one, as the description the caption offers may in some instances be plausible but not necessarily apparent from the photograph. Here, the authority for the caption does not lie within the photograph; rather, the caption shifts attention away from the photograph. However, even if there does not appear an easy and evident connection, the co-presence and proximity of the text and image, together with the markedly different type (size, color) of the text invites reading to produce a relation (Figure 2). The caption of the first image in the feature online text reads, “The tools are 4,300 years old.” The caption says nothing about stones but rather has as subject “the tools,” which are described by the predicate “are 4,300 years old.” But what can be seen in any one display is not self-evident. This as has been clearly shown in an analysis of the court proceedings of the Rodney King case, where a police officer “taught” the jury and judge how it is that Rodney King was not beaten gratuitously but displayed aggressive behavior that required officers to subdue him with force. Here, the definite article “the” constitutes a resource for reading to organize itself to find the determined and therefore determinate objects/categories written about. In the same way as the definite article “the” operates between two parts (types) of text, it also operates between caption and image. Thus, because “the tools” refers to something (supposedly) already known or present, it encourages reading that has not seen them yet to search for “the tools.” Proximity and common hierarchical relations in Western texts allow readers to seek and find the tools in the photographs immediately above the text in the same column. The text “The tools” then functions as a pedagogy of naming accompanied by an index, as if it had said “these are tools” in the way a parent points to an image in a book for toddlers and young children while reading with/to them. The caption then not only provides a description but also, in fact, constitutes an instruction for how to see:

tools rather than stones. It functions precisely like the parent’s finger pointing to the stone tools in the image while uttering, “these are stone tools [chimps use for cracking nuts].” As a description, the caption has been occasioned by the image, which is the what that the description describes. The text is in grey and printed precisely underneath the image, allowing reading to associate it first with the image before identifying or relating it with other text. But being grey and smaller, the text wants to withdraw, allowing predominance to the regular and even more so to the bold-faced text. In its attempt to withdraw into the ground, the caption also invites reading to consider the image in itself. In the caption, too, there are resources that structure reading. Following subject “The tools” and the predicate “are 4,300 years old,” there is a comma followed by the words “say researchers.” The clause is read as modifying the statement that precedes it; but it does so not in a definitive way. One way in which the clause may be read is as saying that “researchers” rather than other individuals / professionals say, which would confer authority to the statement. It also allows reading the sentence as a modifier that weakens the statement from actual fact to something (some) people say. In this way, the clause supports a reading structured by the inverted commas that indicate something as reported speech. What in the statement “the tools are 4,300 years old” may be the contested or contestable part? That the stones are to be seen as tools or their age in years? In fact, the stones themselves are likely older. It is their tool age that could be in question, and this is how we have to read the age to make any sense at all. The tool nature may be seen for those already familiar with some human stone tools, arrow/spear head and a scraping tool (bottom right). However, reading of the image cannot confirm the age of the stone tools. It is impossible to see the age of stone or stone-as-tool; this part of the caption, as indicated above, therefore requires the kind of authentication outside the image. The second image/caption pair presents us with an additional problem that often is present in the online science texts: a gap between what the caption says and what the image supports. Reading encounters the photograph of a chimpanzee (for readers knowing chimpanzees) accompanied by the inscription “The use of stone tools may have a deep evolutionary origin.” In the photograph, we do not see any of the stone tools that are the topic of the article but a member of the species that figures as the subject. In fact, the caption predicatively relates the use of stone tools to evolutionary origins. The only other aspect of the display that falls into the category collection or family with “evolutionary origin” is the category of “common ancestor of chimps and humans.” Here, there is an activity, the use of stone tools, also bound to chimpanzees though normally only bound to humans. The caption invites the extension of the normally category-bound activity of stone tool use to chimpanzees, a member of which appears in the photograph. The caption thereby does not actually describe the contents of the image— stating what would be the obvious for the informed reader, though in school textbooks an inscription such as “A chimpanzee” would have definitive purpose. But the collocation of stone tool use (caption), chimpanzee (image), and the category binding to humans (“tool use”) finds an equivalent concept of “deep evolutionary origin.” Thus, even if unfamiliar with the possibility of “a common ancestor of chimps and humans” (¶3), reading would be configured to anticipate reading about the evolutionary link between chimpanzees and humans. Subtitles In contrast to the title, the boldfaced text immediately below it does not use the same mediating structure (ends with a period “.”) but elaborates the former in several ways. First, in the same way that the caption elaborated the figure, this title below the title concretizes the term “ancient” by providing a number to specify the period of time numerically: 4,300 years ago. It elaborates the chimpanzees by specifying a geographical location where they lived and, therewith, where the research was conducted, or at least, where the artifacts have been found (the age determination using radiocarbon dating probably was not done at the site itself). Finally, it elaborates the objects that the tools were to be used for: the tools were used to crack nuts. How does the text achieve “elaboration,” or rather, how does the text|reading of the first paragraph elaborate the results of the preceding text|reading of the title? First, this text stands between title and the text itself. It is boldfaced thereby distinguishing it from other text. Its size is the same as a subsequent title and as the title of the entire text, printed in larger font size. Second, it is an

elaboration of the title, a fact clearly established in the next statement (retroactively), which specifies the newsworthy item in and of the present page: “the discovery represents the oldest evidence of tool use,” modified by “our closest evolutionary relative.” Only in the next paragraph is the statement categorized as “ the discovery.” All other subtitles, physically set apart as such by the different size and type (boldface), engage in different work, as they always consist of a noun phrase without predicate (here “Nut crunch”). Because it does not make a statement, the noun phrase serves as a sign to look out for what is coming. Reading subsequently discovers what the noun phrase announces: here, two paragraphs in which (a) the stones are linked inferentially to nuts in a complement to the verb and (b) a justification for this inference in identification of starch traces on the stones to locally found nuts. However, the “nut crunch” does not bear any (obvious) relation to the remaining three paragraphs—repeating a standard patterns in this medium— so that the “subtitle” may be more a resource for perceptually structuring the display than serving a conceptual function in the way subtitles do in scientific texts (the “results” section features results, not methods, discussions, or implications). Main Text–Image/Caption While reading the first mutually constitutive {caption | image} pair, two questions may possibly emerge in and from the text|reading pair (prompted by “the author say”): (a) how can one determine that this stones are tools and (b) how can the age be determined to be 4,300 years. Following the boldfaced noun phrase “Nut crunch” (which experienced reading understands to be a “subtitle”), resources are provided to attribute tool nature to the stones. Using the definite article “the” and the subject “excavated stones” the text then provides the predicate “showed hallmarks of use as tools for smashing nuts.” The modifier “when compared with ancient human or modern chimpanzee stone tools” follows. How is it that reading may take the last part of the sentence as a modifier? As competent reading organizes itself moving through the sentence, it can find a sense of completion just after the first appearance “tools,” then again after “nuts,” and again only after the second “tools.” These “hallmarks” are not apparent in the photograph and therefore have to be authenticated, if at all, elsewhere in the text. The second paragraph in this section of the display then provides this authentication: the stones contained starch grains typical of local nuts. A reasonable inference is that their use to crack nuts left the starch grain on these stones. A new question may arise, this time about the identification of nuts as the object worked upon. An answer to the question can and may be found in the next paragraph, which, by means of a repetition of the naming of the activity (“cracking nuts”)—cracking and smashing fall into the same category of verbs. The second question would be answered in the two last paragraphs. In the first of the two, resources in the form of a repetition of the topic “The tools were found to be 4,300 years old” are provided allowing the text/reading pair to establish a relation to the earlier appearing dates (caption, first paragraph). The commas provide resources for reading to produce at least two elaborative clauses. In the first, resources are provided for construing a correspondence with the human stone-age culture: “which, in human terms, corresponds to the later Stone Age.” A second clause adds, “before the advent of agriculture in the area.” An earlier sentence states that the “skill could have been … learnt from humans by imitation.” In this sentence, the stage of human cultural development is characterized as “Stone Age.” The use of stone tools is a category-bound activity: humans living during the Stone Age used stone tools. In this area, Stone Age people used stone tools, thereby providing for the possibility that chimpanzees might have “learnt from humans” using stones as tools.