Slips of the Tongue - A Classification

Table of Contents

1. Introduction

2. Classifying Slip of the Tongue

3. Assemblage Errors
3.1. Anticipations
3.2. Perseverations
3.3. Transpositions and Spoonerisms
3.4. Deletions and Haplologies
3.5. General Remarks Concerning Assemblage Errors

4. Selection Errors
4.1.Semantic Errors
4.2. Substitutions
4.3. Blends
4.4. General Remarks Concerning Selection Errors

5. Conclusion

6. Bibliography

1. Introduction

Whenever people fail to utter the message they wanted to convey correctly, they apologize and tend to explain their accidental mistake as a slip of the tongue. However, what is meant and covered by this term which is used so often to refer to speech errors? Does its universality suggest that all tongue slips are of the same kind? If not, how can they be distinguished?

This term paper will try to give a more detailed categorization of speech anomalies, first providing a definition what is generally seen as a slip of the tongue. It will be shown, who were the first linguists to examine this topic and what discoveries were made.

In order to give an own precise classification of speech errors, different approaches for main categorizations will be presented and taken into consideration. On the basis of these main categories, further sub-categories are elaborated and illustrated with examples given by different authors in order to explain every feature appropriately. Since many authors still not agree upon a common sub-classification, this term paper will select the most suitable terms and definitions, try to compromise about the most common ones and give suggestions for further distinctions. To be conforming to the majority of the literature, the most common terminology shall be used. However, if provided terms are inappropriate or imprecise, they will be modified.

The types of speech errors will be ordered via the main class they belong to. They will be presented as sub-classes, defined, explained and illustrated by examples in detail. Examples are marked with ordinals. The incorrect utterance is always presented with an asterisk and followed by the proper realization in brackets. The pronunciation of the slips is written in graphemes, i.e. normally written words, as far as this is possible to convey the mistake. Where no written equivalent can be found, square brackets will be used for the transcription of the speech error.

After all issues of one main category have been elaborated, rules and general remarks which apply to all of them will be stated.

Finally, this term paper concludes which problems may occur when classifying slips of the tongue and provide an outlook about what speech errors reveal about the speech production process, the inner human lexicon and language universals.

2. Classifying slips of the tongue

Before differentiating slips by their distinctive features, it should be clarified what exactly is meant by the term slip of the tongue. In contrast to the classification, which is debatable and often approached in different ways, scholars broadly agree on a definition. Generally speaking, a slip of the tongue is an anomalous, unintentionally produced utterance. A representative definition is given by Boomer and Laver: “A slip of the tongue is an involuntary deviation in performance from the speaker’s current phonological, grammatical or lexical intention.” (Boomer and Laver 1973: 123)

According to Meringer (1895), the suggestion to examine speech errors can be traced back to Hermann Paul, a German linguist and lexicographer, who claimed that the analysis of slips “might reveal a natural cause of certain types of linguistic change” (Fromkin 1973:13). However, in 1895 Meringer and Mayer were the first linguists to publish a “major psycholinguistic analysis of linguistic errors, together with a corpus of over 8000 illustrative errors” (Fromkin 1980: 1), including anomalies in speech, reading and writing.¹ Their classification covered exchanges, anticipations, postpositions, contaminations and substitutions (cf. Meringer & Mayer 1895). Convinced that people do not misspeak at random , Meringer attempted to present rule-governed connections between the particular slips he had collected, since the regularity in which speech errors occurred implied a constant psychological influence on them. Hence, the analysis of slips would provide an insight into speech mechanisms. Some examples given in Verlesen und Versprechen were reinterpreted by Sigmund Freud in his “classic psychological treatment of speech errors” (Fromkin 1980: 1) by the title Zur Psychopathologie des Alltagslebens, published in 1901. While Meringer had provided an analysis of errors which may be classified as exclusively phonological, “Freud’s interest was in how unconscious ‘slips of the tongue’ reveal underlying, repressed thoughts” (Fromkin 1973: 13). Not only did Freud reuse some of Meringer’s examples without mentioning his sources, but he also stated the main reasons for tongue-slips to be psychological motives rather than phonological word features.

In many ways, the acrimonious disagreements between Meringer and Freud have become a nonissue today among most researchers [...]. Few, if any of us, would argue in favor of Freud’s extreme position that all speech errors, except perhaps for some of the simplest cases [...], could be accounted for by his theory of the unconscious and explained as being caused by repressive mechanisms. But Meringer’s equally extreme Position, which totally discounted any such causes, cannot be easily supported either.

(Fromkin 1980:2)

Although not covering all possible kinds of slips, Meringer’s work provided a rich source with a considerable amount of data for researchers to come; His classification of slips is still effective today and will partially be referred to in this term paper. While scholars agree on the definition of tongue-slips, this is not the case with their subdivision. Therefore, different approaches will be presented in the following.

Slips of the tongue occur in different speech units of varying size. They may affect phonemes, morphemes, whole words or, more rarely, sentence fragments and even whole sentences (cf. Boomer & Laver 1973: 123). As Aitchison points out, phonemes and words are the most commonly affected units (Aitchison 1976: 246). This statement is confirmed by Boomer & Laver whose research results show that “the units most often involved are segments, morphemes and words. Segmental slips are by far the most common, involving about sixty per cent of the examples” (Boomer & Laver 1973: 123). Their classification of speech errors shows “three general modes of slip: Misordering of units in the string, Omission of a unit, or Replacement of a unit” (Boomer & Laver 1973: 123). By stating that tongue-slips are “transient malfunctions [...] of the speech production process which obey stringent linguistic constraints” (Boomer & Laver 1973: 130), Boomer and Laver to a large extent support Meringer’s point of view and concentrate on phonologically explainable issues.

Nooteboom distinguishes between “errors in the programme” (Nooteboom 1969: 146) and “errors of selection” (Nooteboom 1969:154), covering not only wrongly uttered segments (i.e. errors in the program), but also the exchange of a word with a somewhat related, but correctly uttered one (i.e. errors of selection). His term “errors in the programme” covers segmental slips and is divided into “phonemic speech errors” (Nooteboom 1969:146) and “non-phonemic errors” (Nooteboom 1969: 151), the former including the classification given by Meringer, the latter containing “meaningless combinations of phonemes” (Nooteboom 1969: 152), morphemes (including root morphemes and affixes) and whole words (cf. Nooteboom 1969: 152). Stating that distinctive features do not behave like independent elements (cf. Nooteboom 1969: 151), he disagrees with Hocket, who, in contrast, “implies the independence of such features” (Fromkin 1971: 217). Hocket does not refrain from Freud’s point of view concerning anomalous utterances and concentrates on psychological reasons rather than on Meringer’s way of thinking.

Since slips of the tongue demonstrably occur out of several reasons and may not be explained by only one of the two classical approaches, Butterworth appropriately takes both Freud’s and Meringer’s points of view into consideration. He distinguishes between “errors due to alternative expressions of the same intended thought”, “errors due to other elements of an intended utterance” and “errors due to not intended thoughts as proposed by Freud” (Butterworth 1981: 629f.). According to this classification, errors where, for instance, a wrong hyponym is chosen, would fall into the category of “not intended thoughts as proposed by Freud”. However, mistakes of this kind do not necessarily have to occur as a result of suppressed intentions and feelings, but simply because of incorrect selection which cannot be explained by a competing alternative expression. Note that some of Meringer’s subclasses and examples would fall under this category, too. Hence, defining all kinds of selection errors as Freudian slips is not a satisfactory categorization.

When searching for a classification which covers as many different kinds of tongue-slips as possible, Jean Aitchison provides an adequate suggestion:

Broadly speaking, we may categorize the speech errors of normal speakers into two basic types. First, we have those, in which a wrong item is chosen, where something has gone wrong with the selection process. For example:

*Did you remember to buy some toothache? (Did you remember to buy some toothpaste?)

Such errors are perhaps more accurately labelled ‘slips of the brain’.

Second, we find errors in which the correct choice of word has been made, but the utterance has been faultily assembled as in:

*Someone’s been writening threat letters (Someone’s been writing threatening letters.) (Aitchison 1976: 241)

This term paper will take over Aitchison’s distinction and hereinafter refer to the main categories of slips of the tongue as assemblage errors and selection errors. Both terms cover several subclasses of slips. Aitchison provides the following main types:

illustration not visible in this excerpt

Figure 1: The main types of selection and assemblage errors (Aitchison 1976: 246)

This distribution will also be taken into consideration and augmented with further distinctions, in order to provide a more detailed distribution which shall be explained in the following chapters.

As it has already been stated in the last chapter, assemblage errors are slips in which the right utterance is chosen by the speaker, but its phones, syllables or whole words are assembled and ordered somewhat incorrectly. Since these kinds of anomalies are the most frequent ones and occur far more often than selection errors, they shall be examined first.

Concerning assemblage errors, Aitchison gives “three main types: transpositions, anticipations and repetitions” (Aitchison 1976: 244). This classification equals Meringer’s, since the term anticipations is adopted, repetitions mean the same as postpositions and transpositions are exchanges.

Note that among assemblage errors “by far the largest percentage of speech errors of all kinds show [...] [slips] of the size of a phone. These occur both within words and across word boundaries” (Fromkin 1971: 218). Therefore, among all provided examples, the smallest units remain the most slippable ones (cf. Aitchison 1976: 246).

3.1. Anticipations

“Most of the segmental errors are errors of anticipation [...]” (Fromkin 1971: 218). Here, a speaker utters a sound, morpheme or whole word, which is planned to occur in their statement in a latter position, too early (cf. Aitchison 1976: 245). When “a sound several words away is sometimes accidentally activated before it is needed” (Aitchison 1976: 65), there are two types of anticipation according to Meringer (cf. Meringer & Mayer 1895: 28). However, he does not provide a terminology for their distinction. Hence, they will be referred to as replacement and addition. In anticipation with replacement, an item (A) is substituted by another item (B) which should be uttered later on. In addition, no other substitution is produced for the second sound or word (B), so that two similar sounds are produced in one utterance, where the first one (A) has unintentionally been affected by the second one (B), as in an example given by Fromkin (1973: 243):

illustration not visible in this excerpt

Here, the sound [k] in <Canadian> is simply replaced by the [t] in <Toronto>. As shown above, the second item (B), in this case <Toronto> remains unaffected.

In anticipation with addition, the first item (A) is not pushed aside by another item (B), but both items accompany each other in the position of the first item (A), which often results in a contamination of the two and can neither be interpreted as a transposition nor as a blend. Consider, for instance, Fromkin’s example (1973: 246):

illustration not visible in this excerpt

This utterance shows an addition of [r], occurring after a voiced plosive in the second position of an initial consonant cluster in <gift>, which has been affected by [r] following a devoiced plosive in an initial consonant cluster in <Christmas>. If the example above was not anticipation with addition, but rather anticipation with replacement, it would be affected as follows:

illustration not visible in this excerpt

Anticipations with addition are a rather rare phenomenon. Only few examples can be found in the literature:

1)

a) * thrink through (think through)
b) * in order for the stell to form (in order for the cell to form)
c) *a whole blox of flowers (a whole box of flowers)

(Fromkin 1973: 246)

Although the examples given above only show anticipations with addition in initial consonant clusters, this does not have to be a rule. It is rather likely that this type of slip just behaves like anticipation with replacement. The latter may occur in whole words, morphemes and phones within one word or over word boundaries. Consider a slightly changed example given by Aitchison (cf. example 2a) and one given by Meringer (cf. example 2b):

2)

a) *I want to tell Millicent what Millicent said. (I want to tell Mary what Millicent said.) (cf. Aitchison 1976: 245)

b) *Ich gebe mir keinen Witz mehr, über die Witze nachzudenken. (Ich gebe mir keine Mühe mehr, über die Witze nachzudenken) (Meringer & Mayer 1895: 30)

[...]

¹ Meringer and Mayer published their work Versprechen und Verlesen: Eine psychologisch-linguistische Studie (VuV) in Vienna in 1895. It was republished by Cutler and Fay in 1978, who “in their introduction to VuV [...] discuss the ways in which Meringer’s classification and analysis anticipated current views“ (Fromkin 1980: 2).

² „[...] dass die Sprechfehler einen Redemechanismus enthüllen, der die Art, wie entfernte Laute (im Worte oder Satze) aufeinander einwirken, zeigt.“ (Meringer & Mayer 1895: 8f.)

³ Note that in more recent studies on tongue-slips, only those errors are called anticipations in which the first item (A) is substituted by a too early occurring item (B). Additions which may be explained via anticipation, perseveration or other phenomena simply fall into a group referred to as additions. However, in general, made unintentional additions can be identified as resulting from one of the major assemblage errors (i.e. anticipation, perseveration, transposition) or from selection errors, including suppressed thoughts. Therefore, a detailed classification of slips of the tongue should assign additions more precisely to specific error categories, which is attempted in this term paper.