Needs washed

“Its origins need traced.”

(Stabley 1959)

The needs washed construction consists of a form of the verb need (or want or like) followed by a passive participle. For example, in sentence (1), needs repaired is an example of this construction; it has needs as its form of need followed by repaired as its passive participle:

1) The car needs repaired.

In standard English, (1) would not be acceptable. Instead, repaired would either need to become an infinitive (a verb form with to), as in (2a), or a gerund (a verbal noun ending in -ing), as in (2b):

2)  a. The car needs to be repaired.

 b. The car needs repairing.

The most common verb associated with this construction is need. However, Murray and Simon (2002) show that want and like are sometimes possible as well, as in the following examples:

3) Cindy, this one [baby] just woke up and probably wants fed.

4) [The dog] sure does like petted.

This construction is also sometimes referred to as need + V-en (or need/want/like + V-en), where V stands for a verb, and -en indicates that the verb has a passive participle ending (this passive participle ending may be -en, as in ridden, or -ed, as in fixed).

Who says this?

Murray and Simon (2002) describe the rough boundaries as Western Pennsylvania, Eastern Ohio, Northern West Virginia, and Central Indiana. Pockets of speakers may exist in places as far-spread as Kentucky and Illinois. This construction is also attested in Scots English, which might be its historical source.

According to Murray and Simon (1999), the need/want + V-en construction displays sensitivity to no significant sociolinguistic factors other than race, and they say that "white [people] favor the construction significantly more than black" people (pp. 149). Murray and Simon (2002) found that unlike white speakers, virtually no black speakers accept like + V-en.

Syntactic Properties


Murray and Simon (2002) observe the following implicational hierarchy with regard to whether need, want, and like are accepted:

5) need > want > like

In other words, if a speaker can use like in this construction, then that speaker can also use both want and need. If a speaker can use want, then he or she can also use need but may or may not be able to use like. Lastly, if a speaker can use need, then neither of the other verbs are guaranteed to be acceptable, though they may be.

By-phrases allowed

While Brassil (2009) claimed that needs washed sentences are generally incompatible with a by-phrase naming the agent of the participle, Whitman (2010) and Edelstein (2014) have argued that by-phrases are, in fact, possible. The examples in (6) from Whitman (2010) and those in (7) from Edelstein (2014) show some examples of by-phrases occurring with this construction:

6)  a.*The car needs washed, not necessarily by you, but by someone before the weekend.

 b.*I’ve written up the document, but before it goes out, it needs checked by Kim, Alex, and Sandy.

7)  a.*The baby wants cuddled by her mother.

 b.*The soul needs fed by creative, multi-dimensional teaching.

By-phrases have traditionally been used as a diagnostic to differentiate some types of verbs. For example, they are one difference between passives and so-called "middles" like (8):

8) This book reads easily.

Passives allow by-phrases, as in (9a), whereas middles do not, as shown by the unacceptability of (9b):

9)  a. The book is read easily by her.

 b. *The book reads easily by her.

Similarly, by-phrases are also traditionally used to differentiate verbal passives and adjectival passives. For example, (10a), which uses a verbal passive, is acceptable, while (10b), which uses an adjectival passive, is not:

10)  a. The Minotaur is being defeated by Theseus.

 b. *The Minotaur seems defeated by Theseus.

Though verbal passives and adjectival passives appear the same, they are distinguished by various morphological and syntactic facts. For example, only verbal passives may appear after the progressive being, as in (10a), and only adjectival passives can serve as objects of seem, as in (10b). See Bruening (2014) for more distinguishing features.

Despite the difference shown in (10), Bruening (2014) argues that adjectival passives in fact can occur with by-phrases, as in the following piece of dialogue:

11) Invading Commander: I want the treasury left untouched!
      Underling: Untouched by anyone but you, you mean.

If Bruening's argument is accepted, by-phrases are not a useful diagnostic as to whether needs washed sentences involve adjectival passives or verbal passives. However, as discussed in the next section, there are other reasons to think that they involve verbal passives.

No adjectival adverbs

According to Edelstein (2014), adverbs which may precede adjectival passives do not occur naturally with needs washed sentences. For example, well may precede adjectival passives, as shown in (12a) and (12b), but well may not precede the passive participle in the need + V-en construction in (12c) (examples (12a) and (12c) are from Edelstein (2014)):

12)   a.*The well written letter

 b.*The letter seems well written.

 c.*The letter needs well written.

If "needs washed" sentences involved adjectival participles, we would expect that (12c) would be possible. The unacceptability of (12c) therefore suggests that needs washed sentences involve verbal passive participles rather than adjectival passive participles.

Purpose clauses allowed

Purpose infinitives (infinitives in which to can be interpreted as in order to) are possible with the needs washed construction, according to Edelstein (2014), who provides the examples in (13):

13)   a.*The new set still needs washed to kill germs.

 b.*Your brain needs fed to work out.

 c.*He wants cuddled to go to sleep.

Subject not necessarily volitional, even with want or like

Edelstein (2014) also claims that the subject of the construction is not necessarily volitional, even when the verb is want or like, both of which would typically have volitional subjects outside this construction. She cites the sentences in (14) as examples of this construction with non-volitional subjects ((14c) is originally from Murray and Simon (2002:41)):

14)   a.*That Doctor of hers wants reprimanded for missing that one!!

 b.*All kids want told off from time to time.

 c.*[a particular plant...] is easy to grow, except that it ‘likes watered every day’.

In these sentences, it is obvious that it is not the doctor's desire to be reprimanded, or the kids' desire to be told off, and, given the context, (14c) is probably not meant to express the desire of a (perhaps anthropomorphized) plant. In these uses, the meaning of the verb comes quite close to 'need.'

Auxiliary need needn't apply

In the presence of negation, the word need can be an auxiliary or a main verb. When it is an auxiliary, it precedes negation and does not have the suffix -s as in (15a), but when it is a main verb, do is needed and need follows the negation as in (15b):

15)   a.*The car needn't be washed so thoroughly.

 b.*The car doesn't need to be washed so thoroughly.

Edelstein (2014) points out that only the main verb need is possible in the needs washed construction. Modal need is not possible. This is shown by the fact that (16a) is unacceptable while (16b) is acceptable:

16)   a.*The car needn't washed.

 b.*The car doesn't need washed.

Psych verbs possible

According to a study of Pittsburgh speakers by Tenny (1998), the needs washed construction is possible for some speakers with psych verbs&em;which express mental and emotional states/events&em; even when the subject is not a volitional agent. Examples of the needs washed construction with psych verbs are shown in (17):

17)   a. Some people need saddened by tragedy, in order to achieve wisdom.

 b. Nobody needs angered by the truth.

Tenny (1998) argues that the needs washed construction involves an unambiguously verbal passive and cannot be an adjectival passive. Therefore, she claims, the facts of the needs washed construction are only compatible with theories that allow non-agentive psych verbs to form verbal passives. As such, the needs washed construction has the potential to tell us something about language that would have been harder to figure out otherwise.

Needs washed data

(open the map in a new window | see the data in spreadsheet format)

Recent Survey Results

The following map shows results from a recent nationwide survey. It represents speakers' judgments of the sentence Most babies like cuddled. There is some description of the map on the left, as well as a legend in the upper righthand side.

Page contributed by Zach Maher and Jim Wood on June 11, 2011

Updates/revisions: August 22, 2015 (Tom McCoy); June 12, 2018 (Katie Martin)

Please cite this page as: Maher, Zach and Jim Wood. 2011. Needs washed. Yale Grammatical Diversity Project: English in North America. (Available online at Accessed on YYYY-MM-DD). Updated by Tom McCoy (2015) and Katie Martin (2018).


Phenomenon Category: 
Tense, Aspect, Mood
Phenomenon Dialect: 
Appalachian English
Midwestern American English