Uh... well...that was a question that...uh...well, I um, guess I'll...I'll...can you maybe...uh...ask that again?
This is a recreation of me trying to answer a really weird question asked to me by a nice (but possibly from Mars) student in one of my Introduction
to Linguistics courses. It's not that my speech production system failed to execute my speech plan; it's that I never really had much of a plan when
I started speaking. See the difference? Filled pauses like "um" and "uh" and restarts are not errors in speech. The simplest explanation of these phenomena
is that they are what you do when you need more time to figure out what you are going to say, or when you change your mind in the middle of an utterance.
OK, so you may well ask at this point: Having eliminated all that, what is a slip of the tongue? They come in many kinds:
Slips can involve almost anything: single segments (that is, consonants and/or vowels), parts of words, words, phrases or whole sentences.
Let's look at these one at a time.
[TOP]
Phonological Slips: Problems with Sounds
It's possible to make a mistake when speaking that simply
involves a single sound. In fact, up to sixty percent of English slips fit this description. Look at these examples:
- (Spoken by an irritated voter: September 2004, Bowling Green, Ohio)
Utterance: The Gore pampaign
Target: The Gore campaign
- (Spoken by one friend to another. The speaker is my friend, the multitalented and mysterious KH, who is a brilliant shining source of error. Not sure I could live without her! What she meant was roughly: "That was our plan before we got that *&^%$* phone call.")
Utterance: That was the pre-phone flan.
Target: That was the pre-phone plan.
- (Spoken by one rattled spouse to the other)
Utterance: "Don't you fake your shinger at me!"
Target: "Don't you shake your finger at me!"
These poor, frustrated folk all had trouble placing their consonants in the right spots in their utterances. In the first slip above, [p] in the second syllable of "campaign" came forward and replaced the initial consonant. This is called an anticipation because, presumably, it is a result of looking forward in the utterance and getting material before you need it. Anticipations, by the way, are the most common kind of phonological slip in English. Listen to the people
around you for a day or two and you'll almost certainly hear a phonological anticipation or two. Then, you can say, perhaps wisely nodding at the same time, "What a fine example of a phonological antecipation error you have just made, my friend." Come on, do it! You know you want to.
In the second error, the [f] sound in phone replaces the [p] sound in the following word "plan". We say that this [f] sound (which occurred earlier in the utterance) perseverated. This means, it hung around longer than it should have and got reused. The technical term is "phonological perseveration". Beware: don't be confused by the fact that we spell the word "phone" with a "ph". It's pronounced like "f". In speech errors, we are always working with sounds, never with the written forms of words. Yes, people do make errors in writing. Have a look at
writing and typing and
braille in the Slip FAQ.
The third slip above is the most dramatic. Here, the flustered speaker swapped two sounds creating what is commonly called a "spoonerism"--named for the good
Anglican priest and scholar
Rev. Dr. William Archibald Spooner (1844-1930).
In this example, the first sounds of "shake" and "finger" are exchanged or metathesized. Note that the first sound of "shake" is written with two letters, but it is a single sound just like the "ph" above in "phone". (I offer for free here the undisputable fact that "Metathesis" would be a terrific name for your next cat. T. s. Eliot simply overlooked it. Not his fault really.)
These kinds of mistakes give us information about how we assemble the sounds that compose words. Imagine that you have a sort of typewriter in your head. Each key of this typewriter has on it one of the sounds in your language. There is a key for the first sound in "f-inger" and the first sound in "sh-ake" and the first sound in "s-pouse". There is also a key for each different vowel sound. There is a key for the vowel sound in "crook", one for "croak" and one for "crock". When you want to speak a certain word, you have to hit these keys in the right order. To make this easier, keys that you are planning to hit light up. They also stay lit for a while after you hit them. That "lighting up" - called "activation" - makes it likely that any key you hit by accident will be one that is already part of your speaking plan. You might accidentally hit a lit-up key too early (an anticipation) or reuse a key you
didn't mean to reuse (perseveration) or you might switch the order of two keys (a metathesis). It's quite rare to make a mistake that involves a sound that is not part of your plan--that is, to use a key that has not already been lit up for use in speaking. I'm quite certain, for example, that I've said all of the following at one time or another: the copeous production of errors in speech of all kinds being perhaps my only super power.
- speech sperrors
- spreach errors
- each sperrors
But, unless I was also in the same sentence talking about gloves or golf or
glutinous glop, I have probably never said any of these:
- gleech gerrors
- geech glerrors
- eech lerrors
The study of speech errors helped create and support the hypothesis that the activation of consonants and vowels occurs in this way and helps predict what kinds of mistakes are likely. It's the beginning of an explanation of why things happen as they do.
Note that you can also make errors with vowels, or with whole syllables.
Occasionally, albeit less frequently, you might make an error where you put the stress on the wrong syllable in an English word. the fact that this happens relatively rarely is information linguists can use to understand more about how word stress works.
[TOP]
Morphological Slips: Problems with Meaningful Parts of Words
Words in English are often made up of more than one meaningful part. Simple examples are words like "fireplace" and "doorknob". You can easily see that there are two parts of both of these words and that each has a meaning. The same is true of a word like "cats". One part of the word means "that inscrutable feline" and the other part means "plural". We say that all of these words--"fireplace", "doorknob", and "cats"--have two morphemes. When you talk, you can substitute one morpheme for another, leave out or add a morpheme, or simply misplace a morpheme. Here are some examples of morphemic speech errors.
- Utterance: "We're not the only ones with screw looses!"
Fromkin, 1973
(Not only a good speech error, but profoundly and universally true!)
- Utterance: And the world miss outs.
Target: And the world misses out.
- Utterance: I have to untie your shoes.
Target: I have to retie your shoes.
- Utterance: The last time we dram.
(dram rhymes with swam)
Target: The last time we drummed.
It's sort of hard to know what exactly the target was in the first error. It was probably "loose screws" or "screws loose". In either case, you can see that the "s" that belongs on "screw-s" got moved onto "loose".
The second error is similar: the -s (really an -es sequence pronounced more or less as /Iz/), was lost off the end of the verb 'misses' and ended up attaching itself to the word 'out' and getting itself pronounced as a simple /s/.
In the third error, the speaker (a frazzled parent) put the wrong prefix on the verb "tie".
In the forth, the speaker failed to put the right past tense marker on the regular verb "to drum". Instead, she changed the vowel to make it fit one kind of pattern of past tenses in English: swim-swam, run-ran, ring-rang. She did notice her error and corrected this one.
What we learn from errors of this sort is that these meaningful parts of words
...these morphemes, are real building blocks in speech production. In some cases, we assemble words before we utter them, choosing the correct
combinement of roots and prefixes and suffixes as we go.
English has its share of morphological slips, but languages like Turkish and Spanish, where there are almost always more than one morpheme per word, have many more. For more information on Slips in other languages, see
The Slip FAQ
[TOP]
Lexical Slips: Problems with Whole Words
Lexical slips occur when you get the wrong word. They come in three sorts: semantically, phonologically, and environmentally mediated. We'll discuss
these one at a time.
Semantically Mediated Lexical Slips
A semantically mediated lexical slip occurs when one word replaces another and the two are related in meaning. Opposites also commonly substitute for one another. Here are some typical lexical slips.
- Utterance: Honey, did you remember to put the milk back in the oven?
Target: Honey, did you remember to put the milk back in the fridge?
- Utterance: Turn on the air! It's too blasted cold in here.
Target: Turn on the air! It's too blasted hot in here!
In the first error, names for two kitchen appliances get mixed up. In the second, a word replaces its opposite. Both kinds are very common.
Parents are a good source of this kind of lexical error. If you grew up with brothers and sisters, you occasionally heard your mother yell things like
this as her speech production mechanism struggled to come up with the right name for the right child:
- Mary, Laura... Carrie ... GRACE! Get over here right now!
If there are more than two kids, the recitation almost always goes in birth order. This reflects the way the forms are stored and usually recited: not, with apologies to Tommy and Dicky Smothers, who Mom likes best. And yes, the parent always knows which child she or he needs front and center, hence the mounting frustration as the list of wrong names lengthens.
The Dreaded "Sweetheart Slip"
Of course, there's also that terrible moment when you realize that
you have just addressed a new sweetheart by the name of his/her predecessor.
You've done this, admit it, and suffered consequences no doubt.
- Oh, Nellie...uh...Laura, I do love you so!
These are ordinary examples of lexical slips of the tongue. With apologies to Dr. Freud, they don't necessarily mean anything about the inner workings of the subconscious mind. Names for familiar people are certainly semantically related and swap for one another just like the oven and the fridge do. Yes, I have heard of people making the Sweetheart Slip and emerging unscathed by immediately entering into a passionate discussion of speech errors and their genesis. See, linguistics is good for you!
Sometimes, two words get selected instead of one and get blended together. For example, I have two daughters whose names are Claire and Guo. We often hear "Guaire", "Cluo", "Gluo" and other variations at our house.
Here's a very nice word blend, given to us by none other than President George W. Bush. He was probably trying to say "misunderstood" and "underestimated" at the same time.
- They misunderestimated me.
Bentonville, Arkansas; November 6, 2000
President Bush was a pretty good source of speech errors, for which I professionally and sincerely thank him. For a collection of Bushisms, click
here.
Note that many Bushisms are not true speech errors but simply awkward turns of phrase--or things that may have sounded fine in context but look odd in print.
Phonologically Mediated Lexical Slips
Phonologically mediated slips occur when the two words, target and utterance, sound alike but are not necessarily related in meaning. I offer you the classic example from the collection of one of linguistics' best-known speech error researchers, Victoria Fromkin:
- Utterance: White Anglo-Saxon prostitute.
Target: White Anglo-Saxon Protestant.
The two words involved have the same starting sounds, the same vowels, same stress pattern and same number of syllables. These errors are often called "malapropisms."
The term comes from the name of Mrs. Malaprop, a character in Richard Brinsley Sheridan's comedy, The Rivals (1775). For more fun with (primarily intentional) malapropisms see
the Fun-With-Words Malaprop page or check out any of the many fabulous Smothers Brothers skits.
Remember that it only counts as a speech error if the person is aware of both words, knows one should not replace another and is not trying to be funny.
Interestingly, people get a lot more flack for a malapropism than they do for a semantically or environmentally (see below) mediated slip. The (unconscious) supposition is that the speaker did not actually know the right word. Although it is possible that some malapropisms are caused by lexical ignorance or misinformation,
many of them derive from perfectly ordinary misfirings of the speech production system and indicate nothing about education or vocabulary size. It is even conceivable that people with larger vocabularies may make more such errors, since they do in fact have more words to choose from.
Sometimes these two factors--semantic similarity and phonological similarity--gang up together in speech errors. Research by Dr. Zenzie Griffin suggests that
there is more name confusion in a household where the children are called Keith, Kent and Kurt than there would be in a household where the boys are named Thad, Rocko and Ethelbert, which sound nothing alike.
Next time you have to explain the Sweetheart Slip to some irritated partner,
consider whether the names you swapped for one another were also phonologically similar. More arguments on your side!
Environmentally Mediated Lexical Slips
These are errors caused when something you are looking at, or thinking about, interferes in your sentence.
- Utterance: I would like a small fries and a salad.
Target: I would like a small golden violin and a bow.
Yes, this one is mine. I couldn't resist putting this one in even if it is just flat weird. I was telling a story to my daughter. It was late, we were riding on an Amtrak to the middle of nowhere and had been riding for a very, very long time and I was starving, wondering if we'd miss the call to the dining car for lunch and thinking more about food than the pretty little story about musical fairies that I was inventing. These are more like the kind of slips Dr. Freud was talking about. They are reasonably rare, but do occur.
This kind of slip often occurs when something you are looking at, like a passing billboard, pops into a place in your utterance where something else should be.
[TOP]
Syntactic Slips: Problems with Whole Sentences
There are a number of ways things can go wrong with entire sentences. We don't mean sentence fragments, or other kinds of casual speech. We're talking
about unintended utterances. These are not counted as syntactic slips:
- Get 'im!
- uh, Not me! I ain't a-gonna get 'im! You get 'im!
- Naw. Let's just let 'im go then.
- 'Kay.
Ordinary, error free speech is characterized by this sort of truncated talk. Really listen to the people around you, and you will see that conversations really consist of fragmented sentences sprinkled with "ums" and "ers". If you ever read an accurate transcript of a conversation, it looks a little crazy at first. the people seem to be stumbling and bumbling all over the place, horribly inarticulate. Listen to the same stretch of speech, however, and it sounds perfectly reasonable. We simply edit out the restarts, ums and speech errors as we listen. One of the ways to deliberately make someone look bad is to simply write down exactly what he or she says, including all and every small disfluency or error. Mark Liberman, in this
2004 language log post discusses the treatment former President Bush has received as a result of his reputation for speech errors.
Perhaps the easiest kind of syntactic slip to diagnose is a sentence combination error. In these errors, the speaker has two different ways of saying the same thing ready at the same time. Somehow, no decision is made to suppress one and go ahead with the other. The result is a combination of both sentences. Here's an example.
- Utterance: Get these little gnomes off my life!
Target one: Get these little gnomes off my back!
Target two: Get these little gnomes out of my life!
Sometimes, words get switched around:
- where wings take dream.
where dreams take wing.
President George W. Bush
La Crosse, Wisconsin, October 18, 2000
Note that, although the two words "dream" and "wing" switched places, the "-s" ending on the noun stayed put. I like this error a lot, but if President Bush had been reading the speech (which I assume was the case) rather than reciting it, we probably shouldn't count it. Well, probably we shouldn't count it in either case unless we are clear about its context. People do make errors when reading aloud: tune into any live newscast for long enough and you'll hear some. People also make errors with memorized material as any fan of live theater can attest, but these are not precisely the same as those in spontaneous speech.
Chief Justice Roberts made a syntactic slip when he bobbled the word order in reciting his part of the Presidential Oath of Office in 2009. Click here to watch speech error history unfolding live. (Yes, they regave President Obama the Oath of Office a few days later...just to be sure it took!
Again though, if Justice Roberts were reading, we can't be sure what exactly was going on. And we probably shouldn't use it, but I couldn't resist.
Here is one I know was not a result of reading:
- And there will be gnashing and weeping of teeth.
Target: weeping and gnashing of teeth.
Said by my friend PD in a perfectly calm, reasonable voice. (Weeping of teeth? Oh my ... heavens!)
It's sometimes difficult to pin down syntactic slips. It's hard to draw the line between inelegant speech and a genuine slip. Is this a slip?
- Does she have an office hour I could come see her during?
I said this to the secretary of a faculty member at Southern Illinois University in 1992. I wanted desperately to make a good impression on this professor, (the aforementioned historical linguist), but I was nervous. It really does sound awkward, and I really should have been able to do better, but it's hard to pin down exactly what went awry there. Maybe, it doesn't look so bad in print, but out loud it sounded...um... like ... not so fluent.
Here is some speech which has been precisely transcribed with unforgiving exactness!
Can you find the one slip?
-
Okay--that's fine. Now, on the investigation, you know, the Democratic break-in
thing, we're back to the--in the, the problem area because the FBI is not under control, because Gray doesn't exactly know how to control them, and they have, their
investigation is now leading into some productive areas, because they've been able
to trace the money, not through the money itself, but through the bank, you know,
sources--the banker himself. And, and it goes in some directions we don't want it to
go. Ah, also there have been some things, like an informant came in off the street to
the FBI in Miami, who was a photographer or has a friend who is a photographer who
developed some films through this guy, Barker, and the films had pictures of
Democratic National Committee letter head documents and things. So I guess, so it's
things like that that are gonna, that are filtering in. Mitchell came up with
yesterday, and John Dean analyzed very carefully last night and concludes, concurs
now with Mitchell's recommendation that the only way to solve this, and we're set up
beautifully to do it, ah, in that and that...the only network that paid any attention to it last night was NBC...
TRANSCRIPT OF A RECORDING OF A
MEETING BETWEEN PRESIDENT (NIXON)
AND H. R. HALDEMAN IN THE OVAL
OFFICE ON JUNE 23, 1972 FROM
10:04 TO 11: 39 AM --
This is Haldeman speaking.
For access to more of the Nixon transcripts and audio see
http://www.lib.berkeley.edu/MRC/watergate.html.
This stretch of speech, like most, is characterized by restarts and hesitations. The change from "conclude" to "concur" was most likely an error. It would help to hear the audio, to hear if Mr. Haldeman's intonation would tell us if he changed the verb because he felt he had made an error.
With your own utterances, it is easier to tell what is a slip and what isn't. You have access to what you intended to say and can make a more-or-less intuitive determination. You usually feel genuinely surprised to hear what you actually said if it's a slip. Sometimes, you can even detect a slip before you say it. You have a feeling about what you were going to say and you notice that it was not going to be
what you intended to say.
[TOP]