Writing Linguistic Research

Linguistics is the scientific approach to the study of language as a human biological system.  The study of language as such is considered a science in a wide-ranging manner.

Linguistic research can be based on theoretical features of grammar such as word/sentence structure, or on qualitative and quantitative aspects of linguistic data. Topics of research can range from ancient writing systems and historical accounts of ‘dead’ languages to judging the grammaticality of a linguistic component on the basis of intuitions of native speakers.

Across the main subdivisions of linguistic study and research, i.e., phonetics, phonology, morphology, semantics, psycholinguistics, language acquisition, communication disorders, etc., certain lend themselves to a more scientific approach than others.   Since the scope of this discipline has become increasingly broad over time, it is important that the usage of the term science be understood within these contexts.  In very broad terms, science is a process of discovering, observing, and examining phenomena in the natural world which in turn sheds light on the relationships between isolated facts and broad concepts.

Know Your Writing Style

Scientific writing is very different from writing in the humanities. The structure of your work must adhere to certain norms to ensure that you can communicate your research efficiently to a fairly broad audience with clarity and uniformity.  Your writing style, terminology, and methodology must represent the area of linguistics being presented.

Scientific writing generally takes the shape of observing language in some form or another, setting forth a theory or hypothesis based on observations, gathering and classifying data, and concisely and coherently examining and discussing findings.  It is important to remember these facts when writing.

1. Make all assumptions clear
2. Keep your language simple and straightforward
3. Avoid value judgments
4. Construct your argument incrementally
5. Provide robust examples
6. Do not try to prove anything; scientists describe observations. A strong methodology trumps any given desired outcome.

Know Your Readers

Individuals who take the time to read and digest linguistics books and articles are usually highly interested and educated in the discipline and are either using material to build upon or to argue against.  In other words, it’s a tough crowd. However, not every reader will be an expert in your field thus it is always crucial to present basic facts about the linguistic components being studied in order to highlight your knowledge of the subject as well as to use as a reference as you build your argument.

Know Your Theoretical Background

Saussurian Structuralism

Ferdinand Mongin De Saussue

The work of Ferdinand de Saussure is considered to be foundational in the field of linguistics.  Saussurian structuralism has been widely accepted as the theoretical background for linguistic study.  Saussure proposed that the basic element of language is the ‘sign’ which symbolizes its ‘referent.’  The sign can or cannot be related to its referent.  For instance, in the case of representing the word ‘house’ with pronunciation or orthography, the latter has no natural relationship with the meaning of the former.  The symbols (sounds, form) of the word are simply signs leading to the abstract concept of ‘house.’

The ‘sign’ is divided into aspects: the signifier and the signified.  The former is available through the senses.  The signified is the concept or meaning that the signifier holds in the mind. Saussure claimed that these signs are the basic elements of language.

Generative Grammar

One of the most fundamental and crucial aspects of linguistic research is to state and explain theoretical assumptions and orientations on which your work is based.  In the content areas of phonology, morphology, syntax, and semantics, most authors that you will read subscribe to the model of Generative Grammar.

Generative grammar embraces the Saussurian dichotomy between synchronic and diachronic description and analysis. Thus when examining linguistic data it is often important to be familiar with previous or historical forms of the structures being studied.

Generative grammarians claim that children are born possessing intrinsic linguistic properties that reflect the innate nature of human language. As children acquire the rules of their native language at a very young age and without much conscious effort, they learn that which is and is not permitted (language-specific constraints).  In conducting research in the areas of sociolinguistics, psycholinguistics, and language acquisition (L1 and L2) an experimental approach is used which is also based on the principles of generative grammar in that underlying representations will surface according to language-specific rules.

In the more theoretical areas of phonology, morphology, syntax, and semantics underlying forms of language are argued for based on surface forms.  For instance, in phonology, underlying representations are presupposed as the most basic form of a sound/word before any phonological rules have been applied. The phonetic representation is articulated after alternations or transformations take place.  Once a relationship between the underlying and surface forms seems plausible, argumentation is constructed to support the observation.

Useful Terminology

Here are several terms with which every linguistics scholar should become familiar.

Know Your Argument

An argument is named such due to its persuasive nature.  The goal of an argument is to make a point in such a way as to bring along others to come to your conclusion.   Thus it is crucial to set forth logical, coherent points or arguments which are based on well-developed premises or facts.   As you begin to formulate your claims, make certain that you understand exactly what it is you are arguing for.  Make sure that you have robust evidence for each claim you put forth.  And then ask yourself the difficult questions that others may ask, such as “What if…” or “How can one be sure that….”

Know Basic Scientific Method

The understanding and application of the basic scientific method are central to the study of linguistics.

Make an Observation

The first task is to ask a question about an observation you have made based on literature or your own perceptions.  Be careful to not state the obvious.  For instance, you do not want to conduct research on the relationship between speech rate and clear articulation since the outcome can be predicted before any testing is done!

Conduct Preliminary Background Research

Next, it is important to conduct preliminary background research to see whether or not your observation has already been addressed by other researchers, or whether you have something original to contribute to the field.  Perusing academic data bases for current primary sources is an excellent first step to see what kind of information is already been circulated around your topic of interest.

Develop Quantifiable and Falsifiable Hypotheses

Your quantifiable hypothesis is the one which you will be testing whereas the null hypothesis is a way of looking for evidence that will falsify or contradict the former with the goal of keeping you focused. In other words, seeing that one can never prove a theory right or wrong but simply show evidence of support, a null hypothesis helps decide whether or not a hypothesis is capable of being proven false.  In this case, the hypothesis must be reconsidered and either abandoned or refocused.

Test Your Hypothesis

Once you’ve constructed a strong and focused working hypothesis, you’ll begin building an experiment or applying your theory to data with the goal of supporting your claim. When conducting an experiment, you will want to make certain that it is ‘fair.’  In other words, you want your results to reflect an unbiased account of your findings. Be clear about which results you will actually be comparing, how you will be controlling variables[3] and how you will distinguish between what is due to chance and what is statistically significant.

Compare Outcomes

To validate your test results, you should be able to make comparisons between your test results and other similar data, which is often referred to as the control group.  This is the subject group/data that is not manipulated or tested.

Control Variables

All aspects about the experiment must be as similar as possible, except for the variables that are being tested.  For instance, if you are testing the perception of speech signals, which requires careful listening on the part of subject, several elements must be controlled for.  First, you want to be sure that none of your subjects has hearing issues.  You also want to test each subject in identical environments so that background noise does not interfere with the listening task. For more information on this topic click here.

Avoid Bias

All individuals are predisposed towards certain opinions about any body of information with which we come into contact.  These biases are shaped by previous life experiences and are most often under our radar. In other words, we are not even aware that they exist.   Since it is only by being objective that true, valid testing can take place, it is imperative that the researcher be aware of and keep in check any judgments or biases that may skew the experimental process.

Differentiate Chance from Statistical Significance

There is always the risk of ‘contaminating’ data such that the results of an experiment can be skewed.  The larger your sample size, the less likely this will occur.  Use statistics to analyze your raw data. This way you will be able to determine whether or not a difference in results is due to random factors.

Analyze Your Data and Draw a Conclusion

The data that is elicited and gathered for an experiment is referred to as ‘raw data’ which simply means it has not undergone analysis.  It is important to organize and classify raw data meticulously, again, so as not to skew results.  Do not be dismayed if your data do not seem to support your hypothesis.  As you interpret data several options may arise.  For instance, results could support your hypothesis in part, or not at all. Evidence could also cause a revision to the preliminary hypothesis.  Evidence can be inconclusive or lead to the development of an entirely new hypothesis.  Be open-minded.  It is highly suggested to elicit the help of a statistician to verify findings.  Once you have understood the relationship between your hypothesis and the results of the data analysis and are confident that you have some type of support, it is time to move to the next step.

Communicate Your Results

Even though this is perhaps the most complex and time-consuming aspect of conducting research, you can build a strong argument by keeping focused on the following questions.

1. What is the expected observation?
2. What has actually been observed?
3. Do the observations support the expectations?

One of the most important aspects of scientific argumentation is carefully using evidence gathered from data analysis to justify claims made.  You must convince your audience that your evidence is reliable and relevant.

Know How to Make an Outline

As you begin to write, be sure to construct a thorough outline.  This will keep you focused.  Your outline will be shaped by the type of research that is being conducted.  Listed below are the major components to be included in linguistic research in the realms of Core Grammar and Experiments.

Know Thyself

As you begin your research project, make sure you give yourself ample time to accomplish your goals.  We suggest you begin with an outline (which can always be revised) and devise a timeline so you can stay focused on the task before you.

Remember, you may be writing research that will change the future!

Return to Research Tutorials