Unit 5: Mapping tenses

Activity 1: Set theory - Functions

For most students, this introduction to set theory should not be new. However, it is likely that you studied functions and mapping at high school in your mother tongue (which is Japanese for most students enrolled in this course). That means that although the mathematics should be simple, the terminology is likely to be new.

Vist this webpage to learn what bijectives, surjectives and injectives are.

In this unit will we map tenses to grammatical meaning and linguistic models. We will also have a go at mapping tenses in English to those in another language of your choice.

Activity 2: Referring to future events in English

Read.

In this activity we will focus on tenses in English only. One tense may have one or more grammatical meanings. In fact, all grammatical tenses have multiple meanings. One meaning may be expressed by one or more tenses. Let's assume that English has twelve grammatical tenses. This number is contentious, but it is the typical number of tenses introduced in text books designed for learners of English language.

Let's see how many ways there are to describe an event (play football) in the future (tomorrow).

Read these statements which each refer to the future.

1. Fred will play football tomorrow.
2. Fred is going to play football tomorrow.
3. Fred is playing football tomorrow.
4. Fred plays football tomorrow.
5. Fred plans to play football tomorrow.
6. Fred is set to play football tomorrow.

This is not a comprehensive list, but you can see that there are many choices. Given that there are multiple ways to express an action in future time, if you want to be able to use English accuracy and appropriately, it is necessary to differentiate between these choices.

Activity 3: Mapping tenses to time

Complete the quiz on ELMS. This quiz needs you to match each of the twelve grammatical tenses to the most frequent usage. For example, future simple is most commonly used to refer to the "future". It sounds easy, but in fact it is not. If you get full marks on your first attempt consider yourself an "expert". I suspect most people who attempt it will not get full marks, though.

Activity 4: Mapping tenses to meaning

Create a map that shows the how tenses link to future meanings.

This is an overview of the elements that you need to consider when creating your map. Remember one tense may have multiple meanings and meanings may be realized by multiple tenses. To make this task manageable, let's limit the tenses to prsent tenses (present simple, present progressive, present perfect simple and present perfect progressive).

1. Domain: all twelve tenses
2. Arguments: all present tenses
3. Codomain: all types of intentions, plans and predictions
4. Image: timetabled future, arranged future, etc.

Activity 5: Mapping tenses to a one-dimensional linguistic model

A simple model is to map a tense to a timeline. The timeline goes from the past through the present to the future (rather like the arrow of time). Each tense can then be plotted as a point or period of time. This simple visualization is used extensively in grammar books and lessons to help learners of English. Do the timelines shown below help you understand the time reference for each of the tenses?

Activity 6: Mapping tenses to a two-dimensional linguistic model

Let's consider a model of language which takes into account the time of occurence (x axis) and the type of occurence (y axis). The time of occurence can be divided into three zones, namely past, present and future. The types of occurences could be split into four categories, namely: single events, repeated events, continuing events or states. This time the name of a tense could be written in at least one of the (4 x 3) 12 cells.

Create a 3x4 table and write in the names of all simple tenses in the appropriate cells For example, present simple can be written in the present-state cell since "I live in Japan" describes a state that is in the present. (Note: It can also be written in other cells.)

Activity 7: Multi-dimensional models

Watch and listen to this short introduction to visualizing high-dimensional space.

Most of the recent breakthroughs in linguistic modelling involve machine learning. When visualizing language, machine learning is an excellent tool but if the program is GPU-intensive is very troublesome (and expensive) to deploy the progam online as a web app.

Activity 8: Mapping tenses across languages

Read.

When translating between languages it is common to find that an expression in the source language is not present in the target language, or vice-versa. It is also common to find that the source language has only one expression while the target language has multiple ways to express the same meaning, or vice-versa. The table below shows how present progressive tense in Japan is translated into three different tenses in English. In addition, you will notice that in Japanese the subject pronouns are omitted, but in English the pronouns are present.

Activity 9: SpaCy

Watch and listen to a short explanation of one way of extracting verbs and verb phrases using SpaCy and Textacy (in Python).

If you would like to know more about SpaCy, check it out by clicking on the link.

Activity 10: Reading (optional)

This blog post discusses issues with describing the future in English. If you want to understand the differences between the different ways of referring to the future, please check it out.

Activity 11: Useful or useless

Look at the following simple visualization using timelines for English based on one verb "eat". Decide whether it is helpful to learners of English, and consider how could it be improved. Finally, consider how verb groups could be automatically extracted from a text and placed on a timeline.

Knowledge and application

Activity 12: Deducing grammatical meaning

When the tense of the verb and the grammatical class are both known, the grammatical meaning can be deduced. The system first needs to identify tense accurately. It also needs to identify the class of the verb accurately. Using the identified tense and the identified class, the grammatical meaning may be determined. However, given the inherent ambiguity in natural language and the complexity of disambiguating the lexical meaning, identification of both tense and verb class is non-trivial.

Identify the grammatical meaning of these five finite verbs.

1. I live in the dormitory.
2. I am watching television now.
3. I am going to have ramen noodles this evening.
4. I usually go to bed at 1am.
5. However, I will probably go to bed much later today

Submit a pdf via ELMS explaining how can a program:

1. identify verbs?
2. identify grammatical tense?
3. identify verb class?
4. deduce the grammatical meaning?