10. The Distribution of Anaphoric ‘So’ in MICASE

Authors: Jennifer McCormick and Sarah Richardson
Date: January 2006
Download this paper as a PDF file now: Distribution of Anaphoric So (PDF)

---

Kibbitzer 10

The Distribution of Anaphoric ‘So’ in MICASE
Jennifer McCormick and Sarah Richardson

Introduction

There are many uses of the word ‘so’ in English; in fact, ‘so’ is the 12th most frequently occurring word in the MICASE wordlist, a database that consists of spoken academic discourse. However, it placed as the 65th most common word in the Hyland corpus which is composed 24 research articles. From this data it is evident that ‘so’, in its many uses, occurs particularly frequently in spoken English. One use of ‘so’ in English is referred to as the so-called anaphoric _so_—for example, in the phrase I think so. According to John Swales, “the heavy restrictions on anaphoric so suggest that it may constitute “a local grammar”, i.e. a distinctive sub-grammar that is best treated separately from the language as a whole.” This leads one to wonder: In which types of speech events is it most common, and which verbs occur (and do not occur) in this structure.

Uses of So

Of the many occurrences of the word ‘so’, there is Deictic so which indicates degree or modifies manner, for example, “they’ve done so well”. Another occurrence is Discourse Marker so which is the most frequently as a transitional method between topics or speakers, for example, “so most of what we know…” Other uses of ‘so’ are: Initial so which marks the relationship between the main and subordinate clauses, for example, As the temperature increases, so does the humidity.; So as a Connective Adjunct of Reason/Consequence, as seen in the sentence, “There had been a power failure, so all classes had to be cancelled
"; and lastly, Idiomatic so, which often forms expressions referring to approximate measurement, for example, “there were about 60 people or so at the party.” Finally, there is the use of Anaphoric _s_o as a Pro-clause complement with a finite antecedent, for example, “I think so”.

To determine the frequency of anaphoric so a sample of 500 entries was taken from the 16,000 total entries of so in the MICASE database. Of these 500 entries, 20 occurrences of ‘so’ were Deictic, 395 were discourse markers, only 3 were Anaphoric so, and 82 were other uses of ‘so’. Thus, these results would suggest that the frequency of anaphoric so throughout all 16,000 entries would probably average around 1%. This is a rare use then, but still one of considerable interest. (See the pie-chart below.)

Anaphoric so

The linguistic environment in which anaphoric so occurs turns out to be quite restricted. First, it is evident that it cannot be used with all verbs because a speaker cannot say, “*I like so”. In fact, the environment would seem to be restricted to verbs that express mental states such as: imagine, know, believe, seem, appear, afraid, suspect, suggest, think, say, hope, suppose, guess, assume, presume, fear, reckon, tell, gather, and trust. The focus of this study was to see with which of these verbs anaphoric so occurred most frequently and also to see who was using this form of so and in which contexts.

Of the 16,000 entries of so, the following were the most common mental state verbs that occurred with anaphoric so. The only mental state verbs with which it occurred were: think, guess, say, thought, believe, hope and seem. The numerical data is given in the table below. In fact, searches for the occurrence of anaphoric so with the verbs assume, imagine, know, suppose, and suspect yielded no results, suggesting that this usage is even more restricted than many would imagine.

Verb	# of occurrences
Think1	152
Guess	10
Say	10
Thought	6
Believe	4
Hope	3
Seem	2

¹ This includes the three entries of the conditional verb tense, I would think so.

Think so

With the use of anaphoric ‘think so’ being much more common than any other verb in MICASE, it was used as the basis for the remainder of the study. Of the 152 entries of anaphoric ‘think so’ in MICASE there was not a significant difference in the frequency of negated and non-negated entries (80 non-negated and 72 negated). Also, as to be expected, the majority of the entries 140 occurred in 1st person singular; 2nd person singular occurred in 7 entries; 2 occurred in 1st person plural; and 3 occurred as in ellipsis. There were no examples of the 2nd person plural.

Speech Event	# of occurences (per 10,000 words)
Lab	4.9
Interview	3.1
Study Group	2.9
Service Encounter	2.8
Academic Advising	1.7
Small Lecture	1.4
Discussion Section	1.3
Seminar	1.25
Meeting	1.1

The above table shows all the speech event types in which I think so had one or more occurrences per 10,000 words. The table thus suggests that anaphoric so occurs more frequently in informal dialogic speech events, such as study groups.

Translation Equivalents of Anaphoric so in other languages

Finally, when looking at the use of anaphoric so amongst native and non-native speakers of English, it was first necessary to look at the structures that exist in other languages which express the same meaning as ‘I think so’.

French:

Je pense que oui.	Je pense que non.
[I think[I] that yes]	[I think[I] that no]
/I think so./	/I don’t think so./

German:

Ich denke ja.	Ich denke nicht.
[I think[I] yes]	[I think[I] no]
/I think so./	/I don’t think so./

Japanese:

Sō to omou.	Sō to omowanai.
[(About that topic) I think]	[(About that topic) I do not think]
/I think so./	/I don’t think so./

Swedish:

Jag tror det.	Jag tror inte det.
[I think it]	[I think not it]
/I think so./	/I don’t think so./

This use of anaphoric so seems to be rare in other languages. For example, in the languages exemplified above, I think so is typically translated by I think it or I think yes and the opposite for the negated structures. Whether this difference between English and other languages constitutes a problem for non-native speakers of English is a question that needs further research.

Conclusion:

The results of this study show that the use of anaphoric so occurs in a specific environment and is less frequent in comparison to the other uses of so. Of all verbs that express mental condition, ‘think’ most commonly occurs with anaphoric so and its primary occurrence is in events in which speech participants speak in a more informal manner. It is also likely that, while non-native speakers understand the concept of anaphoric so, they may be less inclined to use it when speaking in English if their first language uses a different formula.