How to read a linguistics article

Hey there, and welcome to this video. If you're studying linguistics at university, one thing that you'll need to do very often is to read research articles. That is, papers that have between 20 to 30 pages and that report the results of a linguistic study. Reading that kind of material can be difficult, and it can take up quite a bit of time, especially if you're new to linguistics. There can be lots of technical terms, the writing is usually dense and not very reader-friendly, and maybe the authors used methods that you've never even heard of before. Yeah, if you've clicked on this video, chances are that you currently have a reading assignment, and for the moment, things are not going great. Well, if that's true, you've come to the right place. In this video, I want to give you a few short guidelines that make the process of reading a linguistics article as efficient and hopefully as painless as possible. Now, let's first talk about the reasons for your reading assignment. My students get reading assignments for several different reasons. For example, it could be that they have a class presentation, and so they read the article in order to extract the main points, and then they present those points to the rest of the class. Now, the context for a reading assignment is when you're writing a term paper, and it's your job to compare and to contextualize the findings of several articles. Finally, I often give reading assignments as part of a so-called take-home exam. So, I give students an article together with the exam, and I ask them to explain how the findings of the article relate to issues that we've been talking about in our class. In all of these contexts, the overall goal is actually the same. You want to have a basic understanding of the main issues, and you need to be able to explain those issues in simple language to someone who hasn't read the paper. Yeah, let me make this more explicit. Your goal for reading a linguistics article is that you want to pass what I call the roommate test. Now, what is the roommate test? It goes like this: if you have read and understood a linguistics paper, you should be able to explain the following four points to your roommate. Yeah, so that's the roommate test. As I said, four parts. The first one is this: what did the researchers want to find out? What was their research question? The second part is this one: what did the researchers do to address their research question? What data did they collect? How did they analyze it? Did they measure anything? Did they carry out an experiment? Yeah, this is something that you can explain in relatively simple language to someone who doesn't know the first thing about linguistics. Third, what did the researchers observe? What came out of their measurements or out of their experiment? Yeah, what are their results? And that brings us to the fourth item: what did the researchers conclude on the basis of their results? What answer did they give to their research question? And you see how we're getting back from the fourth point to the first point, the research question, where the circle actually closes. Now, in order to pass the roommate test, you need to be able to present these four items so that your roommate actually understands them, even if he or she is not particularly interested in linguistics. If your roommate doesn't get it, that means it's not good enough yet. Yeah, if your roommate doesn't understand what you're saying, you need to go back to work. Speaking of work, let's get to it. Yeah, we know what the final result should be, but how do we actually get there? What is it that we should do first? How do we get started? Now, I find it helpful to start with a piece of paper where I can put the four items that I want to extract from the text. So, what is the research question? What is the analytical method? What are the results? And what is the final conclusion? If I have all of these on a piece of paper, I can make notes for each of them, and I can start to get a sense of what is important for each single item. Okay, now many papers in linguistics are actually helpful in that they adhere to a structural pattern that is called the IMRAD structure. Now, what does that mean? IMRAD stands for Introduction, Methods, Results, and Discussion, and you notice that these four parts correspond directly to our four items. The introduction should feature the research question. The section about methods should explain how the researchers conducted their analysis. The results section tells us what they observed, and the discussion then moves from the empirical observations to the theoretical conclusions. Now, in theory, that sounds great. Yeah, that is very clear. That should be easy to do. Now, in practice, it's often not that easy. Yeah, there are some problems that we need to talk about. And one problem might actually be that the paper that you are supposed to read may not correspond at all to the IMRAD structure, and it could be a paper that goes from one question to another and then to a third one without ever getting to a method section or a discussion of empirical results. That's terrible news in a way, but it doesn't mean it's a bad paper. It just means that you'll have to read it differently. You have to pay attention to different things. Yeah, but even if the paper adopts the IMRAD structure, which is what I want to focus on here, there are some pitfalls that we need to look out for. Yeah, so let's take this bit by bit. Starting with the introduction, a linguistics article will typically give you quite a bit of background on the topic that the study is on. Yeah, that's the difference between linguistics and, for example, work in the natural sciences. So, it's not unusual to have several pages worth of descriptions of how this topic has been treated in earlier research, what one school of linguists have said, how another school has responded, etc., etc. Yeah, you need to cut through all of that. It's very good to know all of that, but when you present the main points from the paper to your roommate or to your classmates, you want to open with a central research question and not with a long lesson on its theoretical background. Students, especially when they're just starting out, they tend to get overwhelmed by all that background information that is presented in the introduction, and once they get to the presentation of the actual study, they are already overloaded with information. So, an important thing to keep in mind while you're reading the intro section is to distinguish between background information on the one hand and information on the research question on the other. At the end of the intro section, you should have a clear sense of what the research question actually is, and you should be able to formulate it in your own words. That's important. Yeah, it should be a question that a normal person can understand. Think of your roommate. Yeah, now let's assume that you've made it safely to the second part, the method section. Here, there might be a description of a corpus study or an experiment or any kind of research design that the authors use to answer their research question. Again, that sounds pretty straightforward, but it often isn't. The method that the researchers use may be unfamiliar to you. The description of an experiment may be very dense, and there may be terms that are completely new to you. It can be a challenge to work through that, trying to understand what exactly the researchers did. But what you absolutely need to understand is what the researchers measured, what is the data, where does it come from, and what did the researchers do with it. Try and find that information first. Yeah, did the researchers measure frequencies of linguistic elements that they found in a corpus or something that they found in a sample of languages from a database? Did they use sound recordings, and did they measure aspects of pronunciation? Did they conduct an experiment and measure reaction times? You know, that is the kind of information that you should pay particular attention to. Very often, if you know what the actual measurements were, you can rela...

linguistic study reading that kind of material can be difficult and it can take up quite a bit of time especially if you're new to linguistics there can be lots of technical terms the writing is usually dense and not

very reader friendly and maybe the authors used methods that you've never even heard of before yeah if you've clicked on this video chances are that you currently have a reading assignment and for the moment things are not going great well if

that's true you've come to the right place in this video i want to give you a few short guidelines that make the process of reading a linguistics article as efficient and hopefully as painless as possible now let's first talk about the reasons

for your reading assignment my students get reading assignments for several different reasons for example it could be that they have a class presentation and so they read the article in order to extract the main points and then they present those points to

the rest of the class now the context for a reading assignment is when you're writing a term paper and it's your job to compare and to contextualize the findings of several articles finally i often give reading assignments as part

of a so-called take-home exam so i give students an article together with the exam and i ask them to explain how the findings of the article relate to issues that we've been talking about in our class in all of these contexts the overall

goal is actually the same you want to have a basic understanding of the main issues and you need to be able to explain those issues in simple language to someone who hasn't read the paper yeah let me make this more explicit your

goal for reading a linguistics article is that you want to pass what i call the roommate test now what is the roommate test it goes like this if you have read and understood a linguistics paper you should be able to explain the

following four points to your roommate yeah so that's the roommate test as i said four parts the first one is this what did the researchers want to find out what was their research question the second part is this one

what did the researchers do to address their research question what data did they collect how did they analyze it did they measure anything did they carry out an experiment yeah this is something that you can explain in relatively simple language

to someone who doesn't know the first thing about linguistics third what did the researchers observe what came out of their measurements or out of their experiment yeah what are their results and that brings us to the fourth item

what did the researchers conclude on the basis of their results what answer did they give to their research question and you see how we're getting back from the fourth point to the first point the research question where the circle actually closes

now in order to pass the roommate test you need to be able to present these four items so that your roommate actually understands them even if he or she is not particularly interested in linguistics if your roommate doesn't get it that

means it's not good enough yet yeah if your roommate doesn't understand what you're saying you need to go back to work speaking of work let's get to it yeah we know what the final result should be but how do we actually get there what is

it that we should do first how do we get started now i find it helpful to start with a piece of paper where i can put the four items that i want to extract from the text so what is the research question what is

the analytical method uh what are the results and what is the final conclusion if i have all of these on a piece of paper i can make notes for each of them and i can start to get a sense of what

is important for each single item okay now many papers in linguistics are actually helpful in that they adhere to a structural pattern that is called the imrad structure now what does that mean imrad stands for introduction methods results and discussion and you notice

that these four parts correspond directly to our four items the introduction should feature the research question the section about methods should explain how the researchers conducted their analysis the results section tells us what they observed and the discussion then moves from the

empirical observations to the theoretical conclusions now in theory that sounds great yeah that is very clear that should be easy to do now in practice it's often not that easy yeah there are some problems that we need to talk about

and um one problem might actually be that the paper that you are supposed to read may not correspond at all to the imrad structure and it could be a paper that goes from one question to another and then to a

third one without ever getting to a method section or a discussion of empirical results that's terrible news in a way but it doesn't mean it's a bad paper it just means that you'll have to read it differently you have to pay attention

to different things yeah but even if the paper adopts the imright structure which is what i want to focus on here there are some pitfalls that we need to look out for yeah so let's take this bit by bit

starting with the introduction a linguistics article will typically give you quite a bit of background on the topic that the study is on yeah that's the difference between linguistics and for example work in the natural sciences so it's not

unusual to have several pages worth of descriptions of how this topic has been treated in earlier research what one school of linguists have said how another school has responded etc etc yeah you need to cut through all of that it's

very good to know all of that but when you present the main points from the paper to your roommate or to your classmates you want to open with a central research question and not with a long lesson on its

theoretical background students especially when they're just starting out they tend to get overwhelmed by all that background information that is presented in the introduction and once they get to the presentation of the actual study they are already overloaded

with information so important thing to keep in mind while you're reading the intro section distinguish between background information on the one hand and information on the research question on the other at the end of the intro section you

should have a clear sense of what the research question actually is and you should be able to formulate it in your own words that's important yeah it should be a question that a normal person can understand think of your roommate yeah now let's

assume that you've made it safely to the second part the method section here there might be a description of a corpus study or an experiment or any kind of research design that the authors use to answer their research question again

that sounds pretty straightforward but it often isn't the method that the researchers use may be unfamiliar to you the description of an experiment may be very dense and there may be terms that are completely new to you can be a challenge to work through that

trying to understand what exactly the researchers did but what you absolutely need to understand is what the researchers measured what is the data where does it come from and what did the researchers do with it try and find

that information first yeah did the researchers measure frequencies of linguistic elements that they found in a corpus or something that they found in a sample of languages from a database did they use sound recordings and did they measure aspects of pronunciation

did they conduct an experiment and measure reaction times you know that is the kind of information that you should pay particular attention to very often if you know what the actual measurements were you can relate those measurements to the research question and you can

figure out more or less on your own what the remaining parts of the analysis must have been so understanding how the study in your paper actually works that's the most difficult part and the most important part of your

reading assignment yeah so this will be the biggest part of the time that you invest when you're reading that article it will also be the part that is the most difficult to explain to your roommate so be sure to work on this until you get

it exactly right now if everything up to this point has gone according to plan you would proceed to the results section and the main problem that's waiting for you here is the interpretation of tables diagrams statistical results etc etc

now if like me you are not a numbers person this can be horrifying yeah however the good news is that it's not necessary to understand all the little details in order to understand the big picture if you've understood the research

question and you've understood what the researchers measured you should actually be able to work out on your own in your own mind what observations the researchers expected and how those would look in a diagram like a bar chart or a line

graph yeah and the big question is are those expectations borne out by the data or are they not so if there are diagrams in your paper those diagrams will typically visualize the data in order to show whether the researcher's expectations

were right or wrong yeah if the researchers present statistics those statistics will typically be used to show that an expectation is confirmed or refuted with confidence so that the measurements reliably show something that is significant now as in the previous section i would

actually encourage you to work backwards try to find the diagram or the table where the relevant results are shown and then try for yourself to work out whether what you see in that graph or in that table confirms

or refutes the researchers expectations and then work backwards go through the text and try to understand all the remaining aspects in that section right sounds easy huh but trust me there are things to be worked out here budget a little bit of time

for this step now in comparison to the first three steps the fourth one will actually be a walk in the park in the sections of the paper that are dedicated to discussion the researchers will present their main conclusion which is essentially an

answer to their initial research question and that sometimes if you're lucky can be a simple yes or no typically however the final parts of a paper will take the discussion to a more general level so here the researchers might flesh out the

broader implications of their work they might draw comparisons to other studies they might raise new questions that could be asked in light of the conclusions or they might suggest follow-up studies that could tie up some loose ants if you're still paying attention at that

point well good for you but nothing except the general conclusion is going to play a big role for the roommate test so these bits of information are nice to have yeah but you don't need every last bit of it so

once you've read through the paper as a whole it's time to get your notes in order make sure that everything fits together and that everything makes sense to you remember the roommate test yeah you want to have all four parts of the article

summarized in such a way that your roommate can understand what's going on and um well if something doesn't make sense to you it won't make sense to your roommate either right and uh let's be honest your roommate you know he's nice and all

but perhaps he's not the brightest person on the planet so keep it simple research question analytical methods results conclusion you need to be able to explain all of that in simple language so that it makes sense to you

and to others okay all of this was purely abstract let's make it a little more concrete with an actual article and uh there's a paper that i selected for this that has been written by holger diesel and it has the title

iconicity of sequence a corpus based analysis of the positioning of temporal adverbial clauses in english right sounds complex but it's actually very very clear now if you want to do this as an exercise you can download the paper

there's a link below this video and you can do everything that we've discussed up to this point so you would pause this video or shut down your browser entirely and come back to the second half tomorrow or whenever you have finished

holograms paper right so if you want to do this go ahead i will continue now right the first step you remember is the piece of paper with the four sections research question analytical method results conclusion and once we have that

we can focus on the first part of the paper and try to figure out what is actually relevant for us and usually the title is a good indication of what the paper is all about in this case iconicity of sequence

what does that mean well in a nutshell it means that the temporal order of parts in an utterance line up with the temporal order of the events that they describe so when we talk about two events and one

happened before the other iconicity of sequence would mean that the earlier event event a is expressed first and the later event event b is mentioned after that so here's an example of a complex sentence that is in line with iconicity of

sequence yeah if i say do the laundry before you watch netflix event a is supposed to happen before event b and i linguistically express event a before i turn to event b here's another example um after you do the laundry you can

watch netflix yeah you see it's the same principle you do the laundry that's event a you can watch netflix that is event b and the linguistic structure maps onto the temporal sequence of those two events as i conceptualize them okay um

you can memorize iconicity of sequence uh like this for clauses with before and after so in iconicity of sequence the main clause comes before the before clause and it comes after the after clause okay that's just a simple way

of memorizing this um so far i've only shown you examples where iconicity of sequence is actually respected there are other examples where it's not so here's one before you watch netflix do the laundry yeah i can say that it

makes sense but the event that is supposed to come later is expressed first yeah and uh the laundry only comes after uh here's an after clause you can watch netflix after you do the laundry again event b is expressed before event

a both of these versions are okay grammatically yeah both sequences work whether or not they are in line with iconicity of sequence so it's not that one is grammatical and the other one isn't but it's that one is in line with iconicity

of sequence and the other one is not all right that's the background yeah and now we come to hollywood's research question which is actually a very straightforward one when speakers of english form complex sentences with before and with after are they

influenced in any way by iconicity of sequence arguably if we present events in an iconic order that would make it easier for our listeners to understand what we're saying yeah because the conceptualization maps onto the linguistic structure in a

more straightforward way so a more specific version of the research question would be whether it's true that in complex clauses the main clause tends to become tends to come before before clauses and after after clauses and this is actually something

that you can measure and that's what you should put on your note sheet for your roommate yeah so the research question is there iconicity of sequence in complex clauses with before and after that speakers actually produce in real

life now how did hogger study this phenomenon his method involves the use of corpus data specifically he collected from a corpus complex sentences with before clauses and with after clauses and he noted for each example the relative position

of the main clause and the adverbial clause so how often do we get a main clause before a before clause that would be in line with iconicity of sequence and how often do we get the other order and the same for after clauses how often

does the main clause come after the after clause or the other way around that means the measurements for this study are observations from corpus data yeah one observation for each corpus example is or is it not in line with iconicity

of sequence right again find a way of putting that on your note cheat in a way that you can remember and a normal person can understand now a couple of minutes ago i asked you to approach the results section

in such a way that you should in your mind imagine the findings and then check diagrams or tables of results to see if what the researchers found actually corresponds with your expectations and uh in holga's paper there's a table that

you see here on this slide yeah and it's a cross table that lists the linear order of main clause and subordinate clause and it cross tabulates initial and final order with after clauses and before clauses so take a minute to look at the table and

check for yourself whether or not that corresponds to your expectations yeah you can pause the video and i will continue in three two one here we go so what the table shows is that the four clauses have the second uh

column there are highly iconic we find them at the end of the sentence which is in line with iconicity of sequence in 81 cases out of 87 yeah so that is a very very strong tendency with regard to after clauses they are

also more frequently found at the end of a sentence but about one in three examples is actually found sentenced initially so what we're seeing there is that generally adverbial clauses have a tendency to occur at the end of a sentence and with before clauses

iconicity of sequence reinforces this tendency and with after clauses iconicity of sequence has to go against that tendency and that results in a relatively higher proportion of initial adverbial clauses with after okay now in this paper the initial

observation is followed by a more sophisticated analysis that investigates three different factors besides the conceptual temporal order of events and this if you like checks how big the effect of iconicity of sequence is in comparison to other factors such for

example as for example the length of the subordinate clause so when speakers produce utterances with long subordinate clauses those long clauses are typically placed at the end of an utterance there's a famous linguistic principle called the principle of end weight

that explains that so long structures are easier to process if they occur at the end and what counts for us and for the roommate test is that conceptual order that is iconicity has an effect even when these other

factors are taken into account so the main point to note for the results is that iconic order uh has a significant effect it can be observed more often than we would expect by chance even if we account for other intervening

factors all right so that completes your basic summary of the paper and you probably know what time it is now yeah roommate time yeah with your notes try to explain everything here to your roommate will it work well probably not the first time but it

will show you where you still have gaps in your argument so what i would recommend is that you're uh you you read the paper one more time with your notes in hand and you check that your points line up with

what is there in the article all right so that's it for this video i hope this approach makes sense to you and that it makes your life just a tiny bit easier if you have comments or questions leave

them below and i'll try to come up with an answer good luck to all of you with your linguistic projects and i hope to see you soon bye