This is a line of inquiry that gets much interest from participants, like students, and very little funding.
The newspaper article is based on this paper:
Dutch courage? Effects of acute alcohol consumption on self-ratings and observer ratings of foreign language skills.
Conclusions:
Acute alcohol consumption may have beneficial effects on the pronunciation of a foreign language in people who have recently learned that language.
and as such the newspaper is accurate.
This is only moderately surprising, since the effects of alcohol are not always in a linear dose-response pattern. Regarding speech effects alone, there are some prior studies going into the same direction:
The effect of alcohol on speech production:
The present study investigates the effect of acute alcohol consumption on speech in Hungarian subjects. The measures used to reveal these effects were tongue-twisters, which were grouped according to their linguistic features. The number and type of speech errors while uttering the tongue-twisters were compared between intoxicated and sober conditions. The results showed that subjects made more speech errors in alcohol influenced than in sober states in all types of the tongue-twisters except for those using foreign words. Changes in the articulation rate, number of pauses and fundamental frequency were investigated as well. In the intoxicated state, no changes were observed in fundamental frequency and articulation rate, while the number of pauses increased.
One possible explanation is of course that that low doses of alcohol loosen the tongue in general and make it easier for foreign language learners to shed their inhibitions resulting from worries about correct pronunciation – or put simply: to stop overthinking it.
This theoretical model led to Guiora’s most convincing experiment (Guiora et al. 1972b) which involved assessing subjects’ pronunciation of a second language after the ingestion of small amounts of alcohol. Guiora views the lowering of inhibitions via alcohol as a means of “operationally inducing a state of greater permeability of ego boundaries or the ability to partially and temporarily give up one’s separateness of identity’’ (Guiora et al. 1972b:427). [From: Affective factors and the problem of age in second language acquisition (1975) – has several studies more, quote chosen only for psycho-analytical language.]
These research findings continue to pile up:
A large body of previous research has shown that affective variables, including anxiety, attitudes, and motivation, influence language achievement (Gardner, 1985; Gardner & MacIntyre, 1992, 1993a; Skehan, 1989, 1991; Spolsky, 1989). Some of the strongest correlations between affective variables and achievement measures involve anxiety.
Less psychology, however, is involved when looking directly at the processing of words in the brain on a neurological level. There seems to be a certain selectivity for alcohol's effects:
Differential effects of alcohol on the cortical processing of foreign and native language:
On the basis of these results, together with animal data about selective influence of alcohol on the 'younger' functional neuronal systems, it is suggested that in the present experiment alcohol suppressed to a greater degree those neuron populations which subserve the use of English language acquired relatively late in individual development.
tl;dr
It's accurate, credible and in line with much of the work done in this field.
Only one caveat in regard to the phrasing of the question and the parts cited: this is only objectively true for low doses of alcohol. With high doses of alcohol the objectively improved performance starts to diminish, only the subjective improvement remains – for a while.
