Navigation – Plan du site

Reframing treason: metaphors of change and progress in new Labour discourse

Emilie L’Hôte & Maarten Lemmens


This article presents a qualitative and quantitative corpus study based on three new Labour election manifestos (1997, 2001 and 2005), as a preliminary analysis of metaphors of change and progress in new Labour discourse. In addition to providing a detailed description of the concept of change and its versatility, it shows that new Labour has systematically presented itself as the agent of positive change. This is achieved through careful use of various metaphors of time and of the tight link between time metaphors and chain-of-event metaphors. We argue that the reason change is portrayed in such a positive light in new Labour discourse is because it allows for the reframing of the party’s own transformation: it is as desirable as change and the future are, and can no longer be perceived as a case of treason to the ideals of Labour. This study is the first step to a larger analysis on a wider corpus including speeches, theoretical articles as well as manifestos.

Cet article présente une étude de corpus qualitative et quantitative qui se base sur trois manifestes du new Labour Party (1997, 2001 et 2005) pour fournir une analyse préliminaire des métaphores du changement et du progrès dans le discours du parti. Il offre non seulement une description détaillée du concept de changement et de sa versatilité, mais montre aussi comment le new Labour se présente systématiquement comme celui qui amène un changement positif en Grande Bretagne, grâce à l’utilisation de plusieurs métaphores temporelles ainsi que du lien évident entre les métaphores temporelles et celles de la chaîne des événements. On propose d’interpréter ce cadrage positif du changement dans le discours du new Labour comme un moyen de recadrer les transformations qui eurent lieu au sein même du parti. Celles-ci deviennent alors aussi souhaitables que le changement et le futur en général, et ne peuvent plus être perçues comme une trahison envers les idéaux travaillistes. Cette étude est la première étape à une analyse de plus grande envergure sur un corpus de textes comprenant des discours, des articles théoriques et des manifestes du parti.

Haut de page

Texte intégral

1. Introduction

  • 1  We would like to thank Eve Sweetser for her precious remarks on the first draft of this paper, whi (...)

“Forward or back? I can only go one way. I’ve not got a reverse gear” (Tony Blair, in Richard 2004: 258).1

1At the 2003 Labour Party Conference as in most of his speeches, Tony Blair clearly predated Barack Obama in presenting his party as the agent of change in his country. But given the protean nature of the concept of change, the two politicians were able to use it to different ends; New Labour’s distinct goal was to reframe a potentially “treason-like” rightward movement on the political spectrum as a symbol of change and progress for the party and for Britain.

  • 2  Here is the text of the original Clause IV, written by Sydney Webb in 1917 : “to secure for the wo (...)
  • 3  “New Labour defined itself by not being Labour. Issues on which the break could be highlighted wer (...)

2Along with Gordon Brown, Tony Blair is indeed responsible for the renewal of the British Labour Party which occurred in 1994-5, and which was validated by a 1996 manifesto entitled “New Labour New Life for Britain”. Blair and Brown’s most controversial decision was to write a new Clause IV for the party constitution, which no longer mentioned the socialist landmark of common ownership of the means of production.2 The party was thus taking a significant move away from the policies of Keir Hardy, Ramsay MacDonald, or even Neil Kinnock. This move to the right of the British political spectrum triggered a heap of criticism for the architects of new Labour, who were accused of being traitors to the ideals of the party3, of selling « designer socialism » and of relying too heavily on « spin ». But it also – and perhaps paradoxically – led to three consecutive Labour victories in the general elections of 1997, 2001 and 2005.

3“Blair not only killed off any residual fear but also made Labour electable by changing the product on offer to the voters and vigorously marketing it” (Ingham 2003: 6). To use Ingham’s own words, it seems that the new product was well marketed – or rather that the party’s transformation was convincingly framed away from treason thanks to the concepts of change and progress.

4Using the party’s three general election manifestos, the analysis in this paper describes the details of this reframing process and shows how movement away from (Old) Labour is framed as movement towards better Labour.

5Based on careful qualitative and quantitative analysis of an extensive corpus of New Labour texts (see section 2), we detail the framing of new Labour as an agent of positive change. We further show that various uses of metaphors of time yield different conceptions of change, which play different roles in new Labour’s discourse. This leads us to the analysis of new Labour’s use of forward movement as a reframing blend: the reason they need change to be framed in a positive light is because they are ultimately looking to frame their own transformation as a positive evolution, and not as a case of treason to the ideals of Labour. But before going into the analysis itself, it is essential to describe the corpus and methods used.

2. Concepts and methods

2.1. Theoretical framework

6One of the major aims of this article is to bridge the gap between discourse analysis (or discourse studies) and cognitive linguistics. As Van Dijk writes (2007: xxii): “It is within the study of discourse that … an integration of cognitive and interactional approaches is most fruitful. The last decades have seen extraordinary advances in both the study of interaction and that of cognition, and the time has come to integrate these results”. As they have been applied to the field of politics, several cognitive models – such as Conceptual Metaphor Theory (Lakoff & Johnson 2001), Blending Theory (Coulson & Oakley, Fauconnier) and George Lakoff’s work on political discourse, framing and conceptual systems (2002, 2004) – have a clear methodological common ground with the cross discipline of discourse analysis: they share an interest in ‘naturally occurring’ language by real language users, they extend the scope of linguistics far beyond mere grammar issues and they analyse “a vast number of hitherto largely ignored phenomena of language use: coherence, anaphora,… mental models and many other aspects of discourse” (Van Dijk 2007: xxi). All in all, this article shows how cognitive linguistics can provide discourse studies with innovative tools that give more depth to the analysis of texts in general, and of political discourse in particular.

  • 4  Fairclough’s New Labour New Language? (2000), is a notable exception. Based on a significant corpu (...)

7As a corpus-based study of political discourse, this article also aims at fusing qualitative and quantitative analyses, both in cognitive linguistics and discourse studies. Too often still in both fields, actual examples of what is being claimed are few, and the length of corpora remains insufficient.4 Fairclough’s analysis of Blair’s doctrine of the international community (2005), for example, is based on four speeches only. Similarly, Lakoff himself admits to a lack of empirical proof for the two models of moral/political systems he presents in Moral Politics (2002: 158): “… it does not have the degree of confirmation that one would expect of more mature theories”. He also remarks that “[s]urvey research has not yet developed an adequate methodology to test for the presence of complex metaphorical cognitive models such as these” (158). Thus, this article offers one possible method for testing both cognitive and discourse analysis theories against a specific set of carefully collected data.

8While our analyses rely on Conceptual Metaphor Theory as well on Blending Theory, the blends presented in this article are not simply treated as instances of conceptual metaphors. According to Fauconnier & Turner (2002), blending is ubiquitous in our cognitive lives, from dolphin-free tuna to computer desktops to complex metaphorical constructs. Similarly, Coulson & Oakley argue that “… conceptual blending has been invoked to explain everything from perceptual experience, to parsing, to participation in rituals” (2000: 192).

9But precisely because of the very general definition of what blending is, there is a need for a clear rationale in its uses and methods. Indeed, “… it seems rather questionable that the exact same blending processes are at work in the perception of motion as in the comprehension of, say, metaphoric blends” (ibid.). As such, “… researchers in conceptual blending ought to strive for more explicit statements of the grounding principles, and more detailed specifications of the principles that underlie blending analyses” (ibid.). This article chooses to narrow down the seemingly endless possibilities of Blending Theory, and concentrates on its analytical potential for complex structure. It will be used solely to detail two “complex cases of meaning construction” (ibid.) primarily identified as metaphorical but which happen to have a more complex structure; in both cases, mappings between the input spaces are not all metaphorical, and at least one input space in the first blend is already a blend itself (see Section 3.2 and 5.2). Blending allows us to zoom in on two specific points, and clearly provides a more thorough analysis than CMT in complex cases. The two theories can actually be understood as complementing each other rather than being mutually exclusive as far as the purposes of this article are concerned.

2.2. Corpus

2.2.1. Primary corpus

10Our qualitative hypotheses are validated by a quantitative analysis of the three new Labour Manifestos (1997, 2001, 2005). Being public declarations of policy and aims before general elections, they provide a good sample of the discourse of the party; they are also highly informative as to the image projected by the party at a particular point in time. At a later stage, these findings will be tested on a wider corpus also including new Labour speeches and theoretical articles.

  • 5  Word frequencies as computed by WMatrix; see below & Rayson 2009.

11The 1997 Manifesto, “Because Britain deserves better”, is composed of 16,819 words. The 2001 Manifesto, “Ambitions for Britain”, is composed of 26,991 words. The 2005 Manifesto, “Britain forward, not back”, is composed of 22,810 words. All in all, the primary corpus used for this article consists of 66,626 words.5

2.2.2. Secondary corpus and baseline

12The selected new Labour data is compared to the Conservative manifestos of the same period (1997 – 2005). By 1997, the Conservative party had been in power in Britain for 18 years and was new Labour’s main opponent; as such it provides important additional information as to the definition of the new Labour Party. The 1997 Manifesto (“You can only be sure with the Conservatives”) contains 19,887 words, the 2001 Manifesto (“Time to deliver”), 12,281 words, and the 2005 Manifesto (“The British dream”), 6,702 words. In total, our secondary corpus consists of 38,871 words.

13The BNC’s written sample will be used as our baseline corpus. This allows us to identify general tendencies of political discourse, as the primary corpus will be checked against a more “standard” collection of written data. The comparison with the secondary corpus highlights tendencies that are characteristic of new Labour in particular.

2.3 Quantitative analysis and WMatrix

  • 6  UCREL Semantic Analysis System.

14Our analysis combines a microscopic approach with a macroscopic (Biber 1988) approach; while examining “the characteristics of whole texts”, we also focus on particular linguistic features (Rayson 2008: 520). Software support for this method is provided by WMatrix (Rayson 2009), an online tool for corpus analysis and corpus comparison. It allows for corpus annotation using USAS6 (UCREL Semantic Analysis System) and CLAWS, as well as for frequency lists and concordances.

2.3.1. Tagsets

  • 7  See for the entire tagset.
  • 8  As a consequence, some of the keyword analyses in this article display different variants of the s (...)

15The initial semantic tagset for USAS was loosely based on the Longman Lexicon of Contemporary English (McArthur 1981). It was then revised and classified according to 21 “discourse fields” (Archer et al. 2002: 16) which branch out into 232 category labels7 (Archer et al. 2002: 2). Table 1 shows the 21 “discourse fields” at the top of the USAS hierarchy; Figure 1 provides a detailed visual representation of one of these “fields”. This type of analysis provides several advantages over simple word analysis, as testified in the rest of this article. Let us note for the moment that it allows for multiword expressions such as “phrasal verbs (e.g., stubbed out), noun phrases (e.g., riding boots), proper names (e.g., United States of America), true idioms (e.g., living the life of Riley)” (Archer et al. 2002: 1)to be counted together; it also groups together variants within a lemma (change, changing, changed), whereas WMatrix doesn’t allow for isolated lemma frequency counts8 (Rayson 2008).

16However, the USAS semantic tagger still has an error rate of 91%, which means that some of its results should be taken with caution. Our data suggests that its use for the analysis of metaphors may still require some adjustments (cf. Koller et al. 2008 for similar comments), and in our opinion, similar reservations are in order when it is being used as a primary source of quantitative data. However, in this paper, the semantic tagger is used for confirming or broadening elements of analysis which were primarily identified thanks to qualitative and key-word analyses. As such, it is an undeniably reliable tool which is very likely to be refined still in the future.


Table 1. 21 “discourse fields” at the top of the USAS hierarchy


Figure 1. Visual representation of the TIME “field” in USAS

  • 9  For these last two reasons, we have chosen to use small caps to refer to these concepts in our ana (...)

17The definitions of what WMatrix’s semantic tags refer to have actually evolved from “discourse fields” and “category labels” (Archer et al. 2002: 16) to “semantic fields” and “semantic domains” (Rayson 2008), to “semantic concepts” (current version of WMatrix). This paper will use the latter, most recent term for two reasons. First of all, the changing terminology underlines the dynamics of a true reflection on semantic taggers and their proper usage; as such, the term concept seems most accurate for describing the conception and use of USAS and WMatrix. But most importantly, referring to key concepts instead of key domains/fields is more in keeping with the definitions offered by cognitive linguistics in this area. Langacker (1987) describes a concept as a unit of mental representation; Croft & Clausner summarize the distinction between concept and domain as follows: “A concept is a mental unit, a domain is the background knowledge for representing concepts” (1999: 3). As will become clear in the rest of the article, neither WMatrix’s tags nor our analyses refer to basic domains of cognition, i.e., “domains which are footed in fundamental human bodily experiences, such as space, time, various sensations, emotions and perceptions, and certain basic social-interpersonal phenomena” (ibid: 6). However, one should keep in mind that the distinction between a concept and a domain is not etched in stone and that “[t]he nature of the concept-domain relation is such that any concept can in turn function as the domain for other concepts” (ibid: 6). As such, some of the concepts identified in this paper may then serve as part of a domain matrix (ibid: 7) for other concepts.9

2.3.2.Keyword analysis

18One of WMatrix’s strongholds is that it allows for what is called keyword analysis, and extends the method to key grammatical categories and key semantic concepts thanks to the available taggers (Rayson 2003). Each word/semantic tag/POS tag in the primary corpus is compared with its equivalent in the baseline/secondary corpus; then the software evaluates whether the difference between the frequencies in the two corpora is statistically significant or not (see below) and finally reorders the word/tag list according to the statistical score obtained. Hence the words/tags that display the most significant differences in terms of frequency are placed at the top of the list (Rayson 2003). The “positive” keywords/tags in the resulting list occur with unusual frequency in the primary corpus in comparison with the baseline/secondary corpus (a “+” sign is then displayed in the table, meaning that the word/tag is in overuse); the “negative” keywords/tags are on the contrary unusually infrequent in the primary corpus as opposed to the baseline/secondary corpus (this is when a “-“ sign occurs in the table, meaning that the word/tag is in underuse).

19Comparisons between two sets of data and key word/concept analyses provide information that a standard frequency list simply cannot yield (Rayson 2008). The words appearing at the very top of a standard frequency list are usually of no great interest to the ensuing data analysis (Rayson 2008). As illustrated in Table 2, among the top twenty words of our new Labour corpus, only 7 are truly significant (marked in with a *) – the others are mostly articles, conjunctions and prepositions which can only very occasionally yield interesting information in corpus comparisons. Even “have”, which is indicated as relevant here, will have to be studied in more detail at a later stage, as the current frequency table does not automatically discriminate between the auxiliary and the possessive verb. However, only the latter is significant for our purposes.

















we *



will *















our *



with *









have *












new *



more *


Table 2. List of the 20 most frequent words in the new Labour corpus

  • 10  See Rayson (2003) for the detail of the calculation of G2.
  • 11  A code yielding p-values for each LL score will soon be added to WMatrix, thus allowing for more p (...)

20Furthermore, Rayson (2008) remarks that while a standard frequency list may lead to fallacious conclusions regarding some differences in frequencies between two corpora which are not actually statistically significant, a comparison between two frequency lists using likelihood-ratio scores (LL) not only includes normalisation as part of the expected value formula, but also testifies to the relevance of the output. Likelihood ratio (G2) is calculated from the natural log of the ratio of observed and expected frequency10 – in our case, it represent the deviation from the baseline/secondary corpus for each word/tag in the primary corpus. The null hypothesis is that there is no significant difference in word frequency between the primary corpus and the baseline/secondary corpus. As likelihood ratio (G2) and chi-square (χ2) have a similar distribution (Johnson 2008: 164), especially for large corpora (Rayson 2003), a chi-square table provides the critical p-values for WMatrix’s LL scores11 (See Table 3).











Table 3. Chi-square table of critical p-values at 1 degree of freedom

21Rayson (2003) recommends to set the cut-off at LL=6.63/ p=0.01, though he remarks that because of the nature of the corpora, results yielding a p-value inferior to 0.001 and 0.0001 are actually more accurate.

22Let us take for example the word new. It occurs 435 times in the new Labour corpus (relative frequency of 0.65) and 152 times in the Conservative corpus (relative frequency of 0.39). WMatrix yields a LL score of 31.90, which means that p<0.0001 at 1 degree of freedom. Thus we can safely reject the null hypothesis and conclude that the difference in the frequency of new between the new Labour corpus and the Conservative corpus is statistically significant.

23All in all, WMatrix is a very useful software for corpus analysis and statistical comparisons which provides convincing results as far as discourse analysis is concerned.

3. Can things only get better? The framing of new Labour as the agent of positive change

24Bybee (2006) emphasizes the importance of the mind’s response to repetition from a grammatical and lexical point of view. She argues that language “does not have structure a priori, but rather the apparent structure emerges from the repetition of many local events (in this case speech events)” (2006: 714). Similarly, repetition is a powerful tool in political discourse, especially when reframing is at stake. George Lakoff (2006) argues that it actually has the power to change brain patterns. According to neuroscientists, “when a sentence is repeated several times, the neural circuits that compute its meaning are activated repeatedly in the brain. Synapses connecting the neurons in the circuits get stronger and circuits may become permanent” (2006: 10).

25Thus as the concepts of change and progress regularly occur together with that of Labour, they will eventually be tied into your brain without your even being aware of the process: “You may not be aware of the moral or political principles that lie behind the political conclusions that you reach quickly and automatically” (ibid.). The primary corpus definitely makes use of this technique in the characterisation of new Labour as the party of change. In his comparison of the 2001 new Labour and Liberal Democrat manifestos, Rayson (2002) shows that the concepts of future and change are mentioned 375 times by new Labour but only 139 by the Liberal Democrats. This article takes a closer look at how the identification process is achieved in a larger set of data.

3.1 Building a definition for “change” in discourse

  • 12  Throughout this article, a couple of representative examples from the three manifestos will be pro (...)

26Etymologically speaking, the primary meaning of the noun change profiles difference: in late Latin, cambi-um meant “exchange, substitution of one thing for another” (OED). The newness aspect of change is an entailment of its original meaning: if you substitute one thing for another, you get something new. And it is precisely this newness aspect that is profiled in new Labour discourse about change. As is to be expected, the adjective new is ubiquitous:12


… new ways and new ideas to tackle the new issues (1997, p. 4)


… the Labour Government has introduced new rights for people at work and new opportunities for trade unions to represent their members (2005, p. 27)


… our country is on a new course (2001, p. 3)

27 The frequency of change and new in the three corpora are given in Table 4;their uses in the three new Labour (NL) manifestos are significantly higher than in the BNC (LL=606.68, p<0.0001 for new, and LL=74.10, p<0.0001 for change) or in the Conservative (CP) manifestos (LL=31.90, p<0.0001 for new, and LL=19.61, p<0.0001 for change).


Freq. NL


Freq. CP



NL vs. CP


NL vs. CP

Freq. BNC



NL vs. BNC


NL vs. BNC

















Table 4. Frequency comparisons for new and change

28Looking at the concordances (part of which is given in Figure 2 by way of illustration) confirms that, contrary what could be expected given the name new Labour, this significant difference is by no means simply due to the co-occurrence of new and of Labour, of which only 59 can be found throughout the manifestos (out of 435 occurrences of new and 301 of Labour).


Figure 2. Extract from the concordance table for new in new Labour corpus

29Newness is also expressed through words such as renewal or verbs like introduce, create, reform or modernise, as well as adjectives like modern:


We have modernised the Labour Party and we will modernise Britain (1997, p. 4)


… opportunity for all – the modern definition of full employment (2005, p. 17)

30At first glance, a key-word analysis (See Table 5) suggests that the use of newness is more of a general political trend than a true characteristic of new Labour discourse, as it is also very much present in the conservative texts; but it is definitely stronger in the NL corpus. Neither modernise, create, nor modernising show significant frequency differences between the new Labour and the Conservative corpora; introduce is actually in slightly significant overuse in the Conservative corpus (LL=6.87, p<0.01).


Freq. NL


Freq. CP



NL vs. CP


NL vs. CP

Freq. BNC



NL vs. BNC


NL vs. BNC














































30/0.00    0.00



Table 5. Frequency comparisons for newness-related key words

31However, the specificity of new Labour becomes clearer with a key concept analysis (See Table 6). While the use of the concept of future in the new Labour corpus is only significantly different from the baseline (LL=1917.51, p<0.0001), change, newness and beginning are used significantly more by new Labour than by the Conservatives (LLchange= 12.02, p<0.001; LLnewness=48.77, p<0.0001 and LLbeginning=31.72, p<0.0001).


Name (tag)

Freq. NL


Freq. CP



NL vs. CP


NL vs. CP

Freq. BNC



NL vs. BNC


NL vs. BNC

Time: future (T1.1.3)








Time: new and young (T3-)








Change (A2.1+)








Time: beginning (T2+)








Table 6. Frequency comparisons for newness-related key concepts

32The WMatrix concordance table for “Time: Beginning”, of which Figure 3 is an extract, makes it clearer why it is included in the newness-related concepts. Indeed, it includes many words which are clearly “newness-related keywords” for our purpose and whose individual significance in the characterisation of new Labour discourse was unclear (such as introduction or start for example). This example emphasizes the higher accuracy of a key-concept analysis, as a complement for a simple key-word analysis.


Figure 3. Extract from the concordance table for time: beginning in new Labour corpus

33As for the etymologically more basic notion of “difference” that underlies the concept of change, while still activated by words like transform(ation) or different, it has a very limited frequency of occurrence compared to “newness”. Key-word analyses do not yield any significant results, and the noun difference actually only occurs 3 times in the new Labour corpus, always within the phrase make a difference. A key-concept comparison suggests a new Labour overuse of the concept labelled “Comparing: Similar/different”, which is only significant vis-à-vis the baseline (LL=10.48, p<0.001). However, a quick look at the concordance table for this concept (See Figure 4) shows that it systematically occurs as a form of the lemma compare; it is mostly activated to show improvement from 1997 and from the years of Conservative government. Thus even this analysis confirms that it is the newness aspect of change that is predominant in the new Labour corpus.


Figure 4. Concordance table for comparing: similar/different in new Labour corpus

34Actually, New Labour’s choice of profiling newness over difference has to be understood in terms of the Party’s general reframing strategy, as will be detailed in the rest of this article.

35Through this strategy, New Labour definitely places itself within the frame of a culture of the future, where change is desirable and the past is something to get away from. This can be compared to what Robin Lakoff (1991) has observed when opposing ancient Rome (a culture of the past) and contemporary America versus (a culture of the future). Romans saw the past as a golden age that had to be sought after, and aristocratic legitimacy was inherited from heroes: “With no public notion of ‘science’, there was no idea of progress, and religion offered only the golden age in the distant past, nothing to come” (Lakoff, R. 1991: 233). On the other hand, contemporary Americans view the past as something to be rid off and do not wish to leave anything behind as they look towards the future: “For us, there is no past to look back upon: if you want a model, look to yourselves” (Lakoff, R. 1991: 234). While British culture is by no means equivalent to American culture, this tendency to look to the future has clearly become a constant between the two cultures.

36It actually provides the basis for the main construct within the party’s discourse: in the manifestos, the change promoted by new Labour is mostly presented as an instance of progress. There is a quasi-systematic co-occurrence of change with increase and verticality:


Since 1997, average incomes have increased by 19 per cent and the income of a typical family has increased by 17 per cent, with living standards rising in double digits for every part of the income spectrum (2005, p. 11)


We will help you get more out of life (1997, p. 23)

37While terms like improve, better, more, rise and extra are in significant overuse in the new Labour corpus as compared to the baseline, the frequencies are not notably different from those of the Conservative corpus, even for extra (LL=9.11, p<0.11) (See Table 7).


Freq. NL


Freq. CP



NL vs. CP


NL vs. CP

Freq. BNC



NL vs. BNC


NL vs. BNC

















































Table 7. Frequency comparisons for increase-related key words

38Once again, this output can be refined by a key-concept analysis. While the new Labour frequencies for ”Quantities: many/much” are not significantly different from the Conservative frequencies, one of the degrees of ”Money: Affluence” shows slight overuse in the new Labour corpus (LL=8.27, p<0.01).


Name (tag)

Freq. NL


Freq. CP



NL vs. CP


NL vs. CP

Freq. BNC



NL vs. BNC


NL vs. BNC

Quantities: many/much (N5++)








Money: affluence (I.1.1+)








Money: affluence (I.1.1++)








Table 8. Frequency comparisons for increase-related key concepts

39In short, as far as the concepts of increase and quantity are concerned, these results point towards a general trend in political discourse rather than a specificity of that used by new Labour, although the tendency is more acute in the new Labour corpus than in the Conservative corpus. The specificity of new Labour discourse lies in this case in the interaction between the concepts of increase and quantity and those of change and future.

  • 13  While Google is not a reliable source for corpus analysis, it does nonetheless provide interesting (...)

40This association is easy, as the combination of two of our most basic metaphors, namely more is up and good is up, instantly allows for “more is better” to be understood as a value coherent with the corresponding conceptual system13 (Lakoff & Johnson 2003). As a result, the manifestos’ repetitions highlight the positive connotations associated with change and future – with new Labour, “change is up”, “change is more” and as a result “change is better”:


But I believe Britain can and must be better: better schools, better hospitals, better way of tackling crime, … of equipping ourselves for a new world economy (1997, p. 1)

41The effect is strengthened by the frequent use of the metaphor complex entities are vertical structures (Grady 1997):


In our third term we will build new ladders of social mobility and advancement on the firm foundations of stability, investment and growth. (2005, p. 18)


… the chance to build the future properly, to make the second term the basis for a radical programme of British renewal (2001, p. 3)


We must build on the British qualities of inventiveness, creativity and adaptability (1997, p.9)


People… building a better future for themselves and for the country (2005, p. 9)


Prosperity needs to be built from the bottom up (1997, p. 13)


It is as if a glass ceiling has stopped us fulfilling our potential (2001, p. 3)


That is the choice: to make progress or to dismantle the foundations laid (2001, p. 3)

42The metaphor applies to various elements in new Labour discourse (such as prosperity, future, or the economy), sometimes with slightly over-the-top results as in example (9): the social ladder is not only being climbed but being built (and multiplied), and seems to have foundations just like a house. However, all these instances are unified under the rather general metaphor Britain is a construction/a building, which is a sub-case of the original metaphor complex entities are vertical structures. But the country is not presented as a centuries-old mansion full of architectural treasures and beautiful antiques to be preserved, which would have highlighted the importance of British history, for example. On the contrary, the construction metaphor is further specified as Britain is a construction/a building in the process of being rebuilt. The architects of the building map onto the leaders of the country (namely here new Labour leaders); the foundations of the building map onto the basis of the country’s policies (10), but also the country’s resources (11); the builders map onto the people who vote for their leaders (12). In that case, the different parts of the building map onto the various elements to which the complex entities are vertical structures metaphor is being applied, such as prosperity for instance (13). In accordance with the future is up metaphor, the parts of the building that are still to be built are mapped onto the future of the country (10), and the sky is the limit once all the glass ceilings have been smashed (14). The common metaphor of glass ceilings map those onto obstacles to progress as upward movement, which range from Conservative policies to old Labour criticism.

43There are two very interesting entailments to Britain is a building in the process of being rebuilt. First of all, if Britain has not finished building yet, one cannot place a definite judgment on the work of the architects – hence, new Labour implies that it takes time to implement policies and that they should only be judged on what has been achieved, not on what is still in progress. But according to the second entailment, if you do not vote for new Labour, the building is going to be destroyed by the Conservatives (15), leaving nothing in its place; even if we ignore the causative reading of (15), the building will at best sink into deterioration. This leaves the voters with a single choice between progress or destruction, thus omitting many other possibilities (such as another building being built, or the same building being built with a different design). As this analysis shows, new Labour manifestos often choose to highlight certain metaphorical entailments in order to create a clear dichotomy between the Conservative Party and the Labour Party, thus leaving voters with an obvious choice.

3.2 Building a new identity for Labour

44By 2005, the party has been so strongly identified with change that Tony Blair’s introduction to the manifesto actually reads: “You voted for change and gave me the chance to serve” (2005, p. 6). He is establishing a metonymic chain, where the head of a political party stands for the party, and a political party stands for what it is promoting. Hence, the head of new Labour stands for change. The party stands for the idea of renewal and difference, which becomes one of its main characteristics. The manifestos construct a very clear identity for new Labour: it is desirable because it is the party of change and of the future – and future is up and change mostly means progress. As one of their election party broadcasts used to sing in 1997: “Things can only get better”.

45But if the party actually stands for change and progress, why not put a capital letter on the word new in New Labour? Contrary to certain policy names, like the “New Deal” (mentioned in the 2001 manifesto, p. 4,), the adjective is systematically used as a modifier for the party name, not as a capitalised part of it:


This is the foundation stone of new Labour (2005, p. 7)


The reason for having created new Labour is to meet the challenges of a different world (1997, p.1)

46Moreover, the name of the party occurs more often as simply “Labour”, or “the Labour Party”, than as “new Labour”. Out of 301 occurrences of “Labour”, only 59 are modified by the adjective “new”, as illustrated by the selection of the WMatrix concordance table in Figure 5.


Figure 5. Extract from concordance table for Labour in new Labour corpus

47Even the header on the 2005 manifesto still reads simply “The Labour Party”. As we show below, this particular phenomenon can be explained by establishing what and who the party is proposing change from, and how the metonymic understanding of Labour as change can help us characterise them.

48Our contention is that a demonization of the past is a necessary step in the framing of new Labour as the party of change in Britain. Indeed, if the past is evil and not desirable, then change is good and a future orientation of politics is desirable.

Tories are zombies

49The most direct Labour opponent is of course the Conservative party. Hence, if Labour stands for change, then the Conservatives stand for more of the same (“18 years of one-party rule”, 1997, p.1); their policies are not new, they are dogmas:


We have no intention or desire to replace one set of dogmas by another (1997, p. 1)


Labour will never put dogma before children’s education (1997, p. 5)

  • 14  The forward-back dichotomy will be discussed in more detail in the third part of this article.

50But most importantly, if Labour moves forward to the future, the Conservatives move back into the past.14 Note that different consequences were possible, such as the Conservatives proposing an alternative direction for the future, but once again, the interpretation creating the clearest dichotomy has been chosen. Voting for the Conservatives is returning to the past:


There will be no return to the penal tax rates that existed under both Labour and Conservative governments in the 1970s. (1997, p. 10)


There should be no return to the 11-plus. It divides children… (1997, p. 5)


… those who seek to take us back to the divisions of the past (2005, p. 8)

51If Labour stands for change and progress, then Conservatism stands for failure – as progress means achieving your goals, absence of progress means failing to achieve your goals. The manifestos mention “… the most dramatic evidence of economic failure” (1997, p. 1) and describe the Thatcher/Major era as a “tragedy…” (1997, p. 1). The past is not only characterised as negative, but also as filled with devils and demons that have to be banished:


… to reject the false choices of the past (2005, p. 7)


… the bad old days of Tory cuts… (2005, p. 29)


we will put behind us the old arguments that have bedevilled education in this country (1997, p. 5)


In our first term we banished the demons of ten per cent interest rates, mass unemployment… and outside toilets in our schools (2005, p. 6)

  • 15  See Coulson & Oakley (2000, 2006), Fauconnier & Turner (2002) and Grady et al. (1999) for a detail (...)
  • 16  While the blend is entitled “Tories are Zombies”, WE admit to using a very liberal definition of t (...)
  • 17  See similar mappings for the analysis of “My karma ran over my dogma” in Coulson & Oakley (2000).

52A 2001 Election Poster epitomizes new Labour’s demonization of the past in presenting Conservatives as characters from a horror movie back from some mysterious place to use evil powers against innocents (see Figure 6).15 Because of the complexity of meaning construction at work in this poster, this example is particularly suited to a Blending analysis.16 Entitled “The Return of… the Repossessed – No home is safe from spiralling Tory interest rates”, it portrays William Hague (the Conservative leader at the time) and Michael Portillo (the Shadow Chancellor) in their black suits emerging from the past of British politics (Input Space 1) as two “zombies” coming back from the realm of the dead (Input Space 2). The “body snatchers” from the original horror-movie become “home snatchers”, armed with “spiralling interest rates” and ready to take possession of British homes (right above the title the fine print reads: “1,453 homes were snatched weekly in 1991”). One of the most obvious reasons for choosing Blending Theory over Conceptual Metaphor Theory in this analysis is that while Tories are Zombies first appears as a simple case of metaphors, the cross-space mappings at work in a Blending Analysis range from identity mappings (fear IS the political opinion generated by the Tories) and metonymy (the black of the suits is metonymic for death in input space1), to phonology17 (pun on “possession”) and metaphor. Thus the meaning construction presented here is more complex than that of a single instance of metaphor. The emergent structure in the blend not only allows for a pun in the title (possessed by evil spirits/home repossession), but also for various references offering a comment on the political situation in Britain in 2001. It is possible to identify a literary reference to Dostoyevsky’s The Possessed, a Russian novel presenting the ideas and ideology of left-wing idealists as demonic, as well as a denunciation of the conservative establishment’s social and political incompetence, but also an allusion to Freud’s Return of the Repressed. In this respect, a blend originally designed to fight new Labour’s main opponent and to reinforce the demonization of Britain’s political past also activates a highly significant interpretation of Britain’s contemporary political scene.


Figure 6. Blending representation of Tories are zombies

Labour’s own demons

53This brings us to new Labour’s other “demon from the past”, namely the pre-reform Labour Party itself:


And we banished Labour demons too: we showed we could run the economy well, cut crime, and stand up for Britain abroad (2005, p. 6)


New Labour is a party of ideas and ideals but not of outdated ideology (1997, p. 4)

54Logically, if post-reform Labour is the party of the future, then pre-reform Labour is the party of the past as well. However, Labour’s past is less frightening than the Tories’ past; it mostly provides inadequate and antiquated means of reaching similar ideals. Thus, the use of new as a modifier rather than a full part of the name of the party aims at showing that new Labour is a different version of the Labour Party, but that it is still the same party with the same values:


New Labour’s 2005 manifesto applies the unchanging values of our party to the new priorities of the British people (2005, p. 5)

55The emphasis is still on Labour, but future-oriented Labour. This provides one more argument for the pre-eminence of newness over difference in the use of change. By insisting on the fact that something is new and thus (in this general framework) better, you demote the fact that it may also be too different from your previous beliefs and be considered treason; it is not, it is simply renewal: the manifestos mention a “new Clause IV”, a “modernised Party” and state that “… the only way to advance decent progressive values was to develop a new policy agenda” (p. 7, 2005).

56At first glance, ”No change” appears relatively prominent in the new Labour corpus (LL=91.04, p<0.0001 in comparison with the baseline), though its frequency is not significantly different from that in the Conservative corpus (LL=0.07) (See Table 9). But a quick look at the concordance table (See Figure 7) explains this statistical oddity, as almost all the occurrences of this concept refer to some form of stability. Hence, the frequency of the no change concept underlines the efforts of new Labour to add continuity to its depiction of change.


Name (tag)

Freq. NL


Freq. CP



NL vs. CP


NL vs. CP

Freq. BNC



NL vs. BNC


NL vs. BNC

No change (A2.1-)








Table 9. Frequency comparison for no change key-concept


Figure 7. Extract from concordance table for no change concept in new Labour corpus

57Thanks to the intricate relation established in the texts between change, future and progress, all the changes within the Labour Party are framed as newer and better versions of what was there before, as opposed to radical changes signifying the end of pre-reform Labour ideals. At the same time, the rhetoric of past vs. future makes the choice between the Conservatives and new Labour extremely obvious.

4. Change and metaphors of time

  • 18  In most of the examples, time is actually not being directly metaphorized. Most occurrences cited (...)

58The uses of the concept of change in the new Labour manifestos also have to do with some of the most basic metaphors in our conceptual system. Indeed, because of “the complex interaction between spatio-physical experience, the human conceptual system and language use” (Evans & Tyler 2003), English speakers understand time according to two fundamental –by now well-known– metaphors which complement each other: moving time and moving ego (Lakoff & Johnson 2003). The occurrences of change in the manifestos can be divided into two main categories, which we call internal and external change. The fact that both models of change use at least one of the two fundamental time metaphors provides us with a deeper insight into the new Labour worldview.18

4.1 Internal change

59Internal change describes change originating in the party and the country themselves. In this first category, it is they who are the agents of change (as opposed to change coming from outside, e.g., globalization-related change).

60The “internal” side of change uses the moving ego metaphor exclusively:


The choice is to go forward to economic stability, rising prosperity and wider opportunities with new Labour (2005, p. 29)


Britain is on the right track. We will not change direction. Neither will we rest on our laurels (2005, p. 7)


… the route to growth (1997, p. 9)


… we have embarked on a radical programme of reform (2005, p. 6)


… I ask you to continue on this journey with us (2001, p. 5)


Our aim is no less than to set British political life on a new course for the future (1997, p. 1)


… set the underperforming school on a new path (1997, p. 6)

61The agents of change are moving along the line of time in the direction of the future, which is of course in front of them – or as we shall see later in this article, “forward” (30). Hence, this side of change is obviously used to frame change as man-made and controlled. The trajector (TR)’s motion is self-propelled – they can decide to go on or to stop (34). Note that because the time metaphor moving ego and the chain-of-event metaphor following your goals is moving along a path share the same source domain (moving along a path), the TR is not simply moving towards the future but towards its goals or ideals (“… economic stability, rising prosperity and wider opportunities” (30), “growth” (32)). The path is then understood as the actions undertaken to achieve these goals (“a radical programme of reform” (33)). Thus the underlying metaphors at play in the occurrences of internal change are: life is a journey (34), goals are destinations (32), to set up a different goal is to change direction along the path and to bring about change is to move along a path towards a destination. This means that the TR is not only responsible for their own movement, but also for the course that they set for their actions. They can decide to take a new path (35, 36) or to go on in the same direction (31). Because of its sense of agentivity, it is this type of internal change that allows new Labour to present themselves as dynamic and proactive.

  • 19  A car is not the only vehicle to have a reverse gear and a driving seat, but it is clearly the fir (...)

62In all the previous examples, Britain itself is understood as a ship or a car that has to be led to its destination; while the state is a ship metaphor is rather conventional, with consecrated phrases such as “the ship of state”, the state is a car metaphor (as in I’ve not got a reverse gear) is rather novel. Both metaphors are joined under the umbrella of the more fundamental the state is a vehicle, of which the state is a car becomes an elaboration.19 In examples such as I can only go one way, we will not change direction (31) or we have driven fundamental reform in secondary provision (2005, p. 31), the designated driver is clearly the party at the head of the state, or the head of state as a metonymical representation of the party. However, there are also several occurrences of the state is a vehicle when the driver is very clearly the people:


… to put the people themselves in the driving seat of our public services (2005, p. 6)


… a health system with patients in the driving seat (2005, p. 67)

63Note that the metaphor of the car makes perfect sense within the politics of new Labour, who emphasizes individual responsibility within the community: a car gives more responsibility to the individual driver, as opposed to a ship where all the passengers are in the hands of the captain and his crew.

  • 20  Cienki (2005), who provides one of the first empirical tests of the theory, shows for instance tha (...)

64The particular development of the concept of internal change also gives more insight into the new Labour world-view. Indeed, the idea of letting people drive the vehicle of state themselves belongs to what George Lakoff calls the nurturing parent model (2002, 2004). While he concentrates on American politics, his findings can also contribute to making sense of British politics: “Our politics are organized around two opposite and idealized models of the family: the strict father and nurturant parent models” (2004: 39). He further argues that: “The nurturant parent family assumes that the world, despite its dangers and difficulties, is basically good, can be made better, and that it is one’s responsibility to work toward that. Accordingly, children are born good and parents can make them better. Both parents share responsibility for raising the children. Their job is to nurture their children and raise their children to be nurturers” (2004: 39-40, our italics). Because of the very theoretical nature of Lakoff’s book, the claim that new Labour’s worldview is of a nurturant parent kind will need to be confirmed with a wider corpus and a more detailed analysis.20 But even if the following examples are only taken as hints to be later confirmed, they are still worth considering within the general argument of the article.

65A key-concept analysis of the new Labour corpus (see Table 10) corroborates the hypothesis that the new Labour world-view is a nurturing one. There is a significant prominence of education, poverty (Money: Lack) and inclusion. Only the frequency of helping does not appear to be significant in the new Labour corpus as opposed to the Conservative corpus (LL= 5.31). While this may point at a general trend in contemporary political discourse (and to a particular conception of the role of government since WW2), it is also at least partly due to the wide-ranging definition of Helping in WMatrix. Indeed the concordance table for S8+ in the Conservative corpus (see Figure 8) includes occurrences of the noun “benefits” that are actually negating the concept (“getting people off benefits”) and many instances of the noun “services” that have little connection to the nurturing parent world view. Thus a more in-depth study of political world views will require a refined system of key-concept annotations.


Name (tag)

Freq. NL


Freq. CP



NL vs. CP


NL vs. CP

Freq. BNC



NL vs. BNC


NL vs. BNC

Helping (S8+)








Education in general (P1)








Money: Lack (I.1.1-)








Money: Lack (I.1.1--)








Inclusion (A1.8+)








Table 10. Frequency comparisons for nurturance-related key concepts


Figure 8. Extract from concordance table for helping concept in Conservative corpus

66In a new Labour is a nurturant parent metaphor, the nurturant parent maps onto new Labour; the family maps onto Britain and the children map onto British people. Raising your children so that they are able to lead their own life as responsible adults maps onto giving British people enough responsibility for them to have a stake in the decision-making process. Believing that your children are born good and that you can make them better maps onto policies that aim at realising the population’s potential, as in:


… over half a million young people helped into word (2005, p. 12)


… collaborate to help youngsters… achieve their potential (2005, p. 40)

67Once again, the manifestos make some of the entailments of this metaphor as biased as possible in order to create a dichotomy between new Labour and their opponents. If new Labour is helping British people forward so that they can realise their potential, then the Conservatives are holding people back and preventing them from realising their potential:


Or whether we will be held back by the traditional British malaise of restricting life’s great opportunities and blessings to a minority (2001, p. 3)


vested interests that hold people back (1997, p. 4)

68Rather strikingly, similar entailments apply to pre-reform Labour landmarks such as the reliance on benefits (described as a trap preventing you from fulfilling your potential) and the theory of class (which is here a barrier on the path to self-fulfilment):


One million single mothers are trapped on benefits (1997, p. 14)


People freed from the barriers of class… (2005, p. 9)

69We describe New Labour’s model as a “nurturant parent with a twist”. The manifestos still aim at drawing a straight line between new Labour as nurturant parent and the Conservative Party as strict father, as in:


… to equip people to succeed… rather than the old Tory way of walking away leaving people unaided to face change (2005, p. 18)

70Example (45) is a clear instance of George Lakoff’s strict father model: “The strict father model assumes that the world is and always will be dangerous and difficult, and that children are born bad and must be made good… Once grown, the self-reliant, disciplined children are on their own, and the father is not to meddle in their lives” (2004: 40-1, our italics). But –and this is where the twist comes in– at times new Labour adopts a very Tory-like point of view in their description of pre-reform Labour – to some the epitome of the nurturant parent model – as a “nanny state”:


Our third-term commitment – not a nanny state but a family-friendly government (2005, p. 75)

71To understand the “twist” in our model, we need to look in more detail at the second, external aspect of change at work in the manifestos.

4.2 External change

72As opposed to internal (man-made) change, external change describes change originating outside the party and the British people; it is framed as an independent entity and thus it comes as no surprise that some instances of external change make us of the moving time metaphor:


Given the pace of change within medical services we will ensure that it is possible for the NHS to change the way in which it organizes its services as quickly as possible (2005, p. 61)


In a fast changing global economy, government cannot postpone or prevent change (2005, p. 18)

73In these cases, change itself is a self-propelled trajector distinct from us. We are still facing the future, but events – and thus change – come from the future and move towards us. As a result, it is not possible to predict the pace of change – we actually have to adapt to it (47). However, because external change is self-propelled, independent and unpredictable, one of the entailments here is that the government cannot be held responsible for whatever comes their way (48), though they have to face it. As was the case with Britain is a construction in the process of being built, new Labour tones down their political accountability by profiling agent-less metaphors (cf. infra).

74But external change is not necessarily a self-propelled trajector moving towards us. It can also be the landmark for another trajector (namely the Party/the British people in the following example):


… we approach new challenges with a progressive strategy for growth (2005, p. 15)

75While here the moving ego metaphor is still prominent, the identities of the landmark and the trajector are more ambiguous in the following instances:


... to govern in a way… that unites our nation in facing the rough and dangerous challenges of the new economy and changed society in which we must live (1997, p. 1)


We do not duck the tough choices (2005, p. 9)

76The reason why these various instances of external change are actually coherent with each other has to do with the orientation of change. Whether it is a trajector (47, 48) or a landmark (49 – 51), external change has a front-back orientation facing towards the past on a timeline. As we are always facing the future, we are literally facing external change (49) whether it is self-propelled or not. Once again, external change being an entity over which the Party has no control, it is harder to hold them responsible for it.

5. New Labour’s “forward” blend

77We have shown how new Labour managed to present themselves as the party of change in Britain, and how their coherent use of two different aspects of change allowed them to build a new Labour worldview – embracing market economy and adding a twist to the nurturant parent model. But the party’s heavy reliance on images of newness and change are actually a way of reframing a more controversial change – that of the liberalisation of the party’s policies. Let us finally concentrate on one of the party’s favourite mottos – “forward, not back” – which we analyse as a blend reframing the party’s rightward movement on the British political spectrum as progress.

78At first glance, the key-word analysis for forward contradicts this hypothesis (see Table 11). Indeed, if the new Labour overuse of the term is significant with regards to the baseline, it is not when compared to the Conservative corpus. However, the concordance tables for the two corpora show that new Labour and the Conservative party use forward differently (See Figure 9 & 10). Indeed, all 7 Conservative instances of the term, as opposed to only 4 out of 22 in the new Labour corpus, concern the phrases push forward, carry forward and bring forward, which denote presentation and realisation; the other 18 new Labour instances clearly profile movement and futurity in the discourse, a sense that is absent from the Conservative corpus as far as forward is concerned. Thus despite a lack of significance in the frequency comparison for forward, new Labour’s use of the term is still highly characteristic.


Freq. NL


Freq. CP



NL vs. CP


NL vs. CP

Freq. BNC



NL vs. BNC


NL vs. BNC









Table 11. Frequency comparisons for forward


Figure 9. Concordance table for forward in the new Labour corpus


Figure 10. Concordance table for forward in the Conservative corpus

5.1 Forward, not back

79The concept of forward movement ends up being a metonymical representation of the party itself, or at least of its program: one element in a political program stands for the whole political program / one element in a political program stands for the political party promoting it. The metonymic relation between new Labour and forward movement is rather obvious in the most straightforward of new Labour’s 2005 Election posters. On a black background, you can see two large arrows pointing in opposite directions: a red arrow that reads “Forward with Blair & Brown” pointing right (i.e., future-oriented), and a blue arrow that reads “Back with Howard & Letwin” pointing left (i.e., past-oriented). Presented in such schematic oppositions (Labour vs. Tories, red vs. blue, forward vs. back), the choice seems easy. But there is a lot more to be said for new Labour’s use of forward movement:


The choice for 2010 is forward with new Labour (2005, p. 41)


It is time for this country to move on and move forward (1997, p. 2)


Our direction and destination are clear (1997, p. 3)


No going back to fiscal irresponsibility. No going back to a Conservative government that says mass unemployment is a price worth paying. Going forward instead to an opportunity economy, spreading prosperity through high employment and welfare reform… (2005, p. 8)


The choice is to go forward to economic stability, rising prosperity and wider opportunities with new Labour; Or go back to the bad old days of Tory cuts, insecurity and instability. (2005, p. 29)


The choice is forward with new Labour to a universal, affordable, good-quality childcare, a million more homeowners, more choice for all parents and an end to child poverty. Or back to the risky economic policies of a Tory government that would let families sink or swim whatever the pressures they face (2005, p. 81)

80Forward movement belongs to the frame of internal change. Once again, it is in turn the country (53), the party (54) or the people (56) who initiate the motion. All these different agents are related metonymically, so that the general concept of forward movement is still coherent. Ultimately, it is British people who are moving forward, with new Labour leading the way: the country stands for its countrymen, the party leading the country stands for the country and the party stands for the people voting for it.

81As previously mentioned, forward movement combines three metaphors: actions with purposeful goals is forward movement, progress is forward and the future is forward. Thus if new Labour stands for forward movement, the dichotomy of entailments is once again present:


The Tories are all talk and no action on family policy p. 81 (2005)


The Conservatives always look back p. 5 (2001)


… those who seek to take us back to the divisions of the past p. 8 (2005)


… a throwback to the 1980s p. 5 (2001)

82Either the Tories are standing still (58), or they are oriented towards the past (59) – if any movement is initiated on their part, it is backward movement, which means a return to the past (60, 61), characterised negatively in the manifestos as shown in section 3.2 above.

83However, the directional dichotomy in the manifestos is cleverly biased. In (55), (56) and (57) for example, Labour and Tory destinations do not belong to the same categories. Labour destinations – when they are actually stated (unlike in (53) and (54)) – are clearly the ideals of the party: “an opportunity economy, spreading prosperity through high employment and welfare reform”, “economic stability, rising prosperity and wider opportunities”, “a universal, affordable, good-quality childcare, a million more homeowners, more choice for all parents and an end to child poverty”. On the other hand, Tory destinations are mostly concrete results and policies as analysed from a new Labour point of view: “fiscal irresponsibility”, “mass unemployment is a price worth paying”, “cuts, insecurity and instability”, “risky economic policies”, “let families sink or swim whatever the pressures they face”.

  • 21  The past has already happened whereas the future hasn’t – thus there are concrete elements to show (...)

84This bias –which is viable because of obvious differences between the past and the future21–is used to buttress the original dichotomy between Tories and Labour, blue and red, back and forward. Indeed, who would be in favour of insecurity and instability, or of letting families sink under pressure? On the other hand, who could oppose high employment, economic stability, or the end of child poverty?

5.2 The Forward blend

  • 22  This article extends Evans & Tyler’s analysis of the semantics of English prepositions (2003) to t (...)

85We have established that forward movement is characterised in a highly positive light throughout the three manifestos. First of all, new Labour chose to identify with the concepts of change and progress because they are very protean concepts that apply to a vast array of situations. But the semantics of the adverb forward show that the insistence on forward movement in new Labour manifestos is also used to reframe another type of – more controversial – movement in the party’s policies, namely its liberalisation.22

86The semantic network of “forward” is grounded spatially: it describes a movement oriented towards the front of the trajector. What is forward “lies in the direction towards which one is moving.” (OED). Hence, a primary scene for “forward” involves a trajector (TR) in motion – its front-back orientation is either inherent (e.g., a person) or is acquired through the motion itself (e.g., a flying rock). The direction of motion is determined by the orientation of the TR’s front. We argue that “forward” is actually determined from the TR’s viewpoint or its intrinsic orientation (for Britain is a car, for example) and does not refer to a fixed direction (See Figure 11). It is this directional versatility that ultimately allows for the existence of new Labour’s forward blend – “forward” can only stand for the direction in which the TR is going.


Figure 11. Viewpoints / intrinsic orientations of the TRs in the orientation of “forward”

87In our moving ego conception of time, we are future-oriented trajectors; so forward spatial motion can easily acquire a temporal sense: forward is future. Note that the versatility of the adverb is already very clear here. Take the following sentence: Next Wednesday’s meeting has been moved forward two days (Boroditsy & Ramsar 2002). It gets different interpretations according to people’s main conception of the time sequence. With moving ego, the meeting takes place Friday, but with moving time, the meeting is on Monday (the TR becomes the event, and not the speaker – hence, forward no longer refers to a later time in the future, but to an earlier time for the speaker). Then in a third sense, we go from temporal orientation to goal orientation (life is a journey and purposeful actions are goals along a path): once again, it is the goal of the TR that determines what metaphorical orientation forward actually takes.

  • 23  Most of the emergent structure of the blend has already been analysed throughout this paper (espec (...)

88Let us now go back to forward movement in new Labour manifestos, which allows for a reframing of the party’s rightward move on the political spectrum in terms of change and progress. Indeed, our analysis of the semantic network of the adverb has shown that the direction that constitutes forward is entirely determined by the viewpoint of new Labour itself. Here is the detail of the blending process at work (see Figure 12 for a visual representation of the blend). Once again, Blending Theory was chosen over Conceptual Metaphor Theory here because of the complexity of meaning construction under scrutiny. As was the case for the Zombie blend, there are different types of cross-space mappings, and the construct cannot be simply analysed as an instance of conceptual metaphor. Input Space 1 (IS1) is change and future for Britain. Input Space 2 (IS2) is liberalisation of the party. Forward movement in time maps onto rightward movement on the political spectrum (from a new Labour point of view). Labour ideals as goals are present in both input spaces – they are linked by an identity mapping. New policies map onto market-oriented policies thanks to a metonymic mapping (one aspect of the policy stands for the policy). The same goes for new ideas mapping onto the absence of clause IV in the constitution of the party. The rupture from the demons of the past maps metaphorically onto the rupture from pre-reform Labour. The crucial point in this blend is that anything that does not originate from a new Labour point of view in the input spaces is not transferred into the blend: there is no place for treason and mixed feelings in the forward blend. Eventually, once the blend is run, the liberalisation of the British Labour Party is entirely framed in terms of change and forward movement.23


Figure 12. Representation of the Forward Blend

6. Conclusion

89In conclusion, this paper has shown how metaphors of change and progress are an essential component of new Labour’s reframing strategy in their election manifestos. The qualitative analysis of the texts using the theories of Metaphor and Blending was combined with a quantitative analysis using the software WMatrix. This allows for an increased accuracy of the results of this study, and is one of its important innovations. A clear profiling of the “newness” aspect of change allows for a positive interpretation of the concept. Thanks to the distinction between internal and external change, change is not only presented as desirable but also necessary, if not inevitable. This allows for the use of the themes of change and progress in the reframing of new Labour’s liberalisation in a positive light. After the “forward” blend, less Labour actually turns out to be better Labour.

  • 24  Comment by “MeandYou” on August 29 2008 to a Guardian article entitled: “In praise of… American rh (...)

90New Labour was not the first party to make use of change, and will certainly not be the last. Barack Obama’s recent campaign comes to mind, of course; he was presenting himself as the candidate of change. Here is a comment from an article in the Guardian, describing Obama’s promises in terms highly reminiscent of new Labour discourse: “Bill Clinton moved US 8 steps forward. George Bush moved the US more than 16 steps backward. You have an Obama who wants to move America forward again and a McCain that promises the same George Bush thing”.24Obama was not using the same forward blend as new Labour, having no need to argue for or against the liberalization of his party. However, it is interesting to notice that this comment, made by a British reader, still uses new Labour terms and dichotomies although the idea of literally ‘moving America forward’ was not as ubiquitous in Obama’s rhetoric as it was in Blair’s.

  • 25  “Israeli Candidate Borrows a (Web) Page From Obama”, by Ethan Bronner and Noam Cohen, The New York (...)

91All in all, this paper’s double goal, viz. of bridging the gap between cognitive linguistics and discourse studies on the one hand, and between qualitative and quantitative analyses on the other, has inspired more reflection on research methods in both fields. On a more practical note, it has also shown how it is the very protean nature of the concept of change that allows for such political framing and reframing to take place. So, it seems appropriate to end with this quote from Ron Dermer, one of Mr. Netanyahu’s top campaign advisers in Israel, commenting on the fact that the candidate’s team blatantly plagiarized Obama’s campaign, thus demonstrating what a vast array of agendas change has the capacity to accommodate: “We’re all in the same business, so we took a close look at a guy who has been the most successful and tried to learn from him. And while we will not use the word ‘change’ in the same way in our campaign, we believe Netanyahu is the real candidate of change for Israel”.25

Haut de page


Archer, D., A. Wilson & P. Rayson 2002. “Introduction to the USAS category system”.

Beech M. & K. Hickson 2007. Labour’s thinkers : the intellectual roots of Labour from Tawney to Gordon Brown,International library of political studies 13. London: Tauris Academic Studies.

Biber, D. 1988. Variation across speech and writing. Cambridge: Cambridge University Press.

Bybee, J. 2006. “From usage to grammar: the mind’s response to repetition”. Language, Volume 82: 711-733.

Cienki, A. 2005. “Metaphor in the ‘Strict Father’ and ‘Nurturant Parent’ cognitive models: Theoretical issues raised in an empirical study” in Cognitive Linguistics 16: 279-312.

Coulson, S. & T. Oakley. 2006. “Purple persuasion: deliberative rhetoric and conceptual blending”. In Luchjenbroers (ed.) Cognitive linguistics investigations: across languages, fields and philosophical boundaries. Amsterdam; Philadelphia: John Benjamins.

Coulson, S. & T. Oakley. 2000. “Blending Basics”. Cognitive Linguistics 11: 175-196.

Croft, W. & T. Clausner. 1999. “Domains and image schemas”, in Cognitive linguistics 10-1: 1-31.

Delaney, P. & M. Emanatian. 1993. “‘Unnatural barriers’: Why metaphor matters (or, linguistics meets the geopolitics of law)”. In De Stadler, L. & C. Eyrich (eds.) Issues in Cognitive Linguistics (1993 Proceedings of the ICLC). Berlin: Mouton de Gruyter, 331-345.

Evans, V. & A. Tyler. 2003. The Semantics of English Prepositions: spatial scenes, embodied meaning and cognition. Cambridge, U.K.; New York: Cambridge University Press.

Fauconnier, G. & M. Turner. 2002. The Way We Think: Conceptual Blending and the Mind’s Hidden Complexities. New York: Basic Books.

Grady, J., T. Oakley & S. Coulson. 1999. “Blending and metaphor”. In Gibbs, R. & G. Steen (eds.) Metaphor in cognitive linguistics: selected papers from the fifth international Cognitive Linguistics Conference, Amsterdam, July 1997. Amsterdam/Philadelphia: John Benjamins.

Grady, J. 1997. “Theories are buildings revisited”. Cognitive Linguistics 8: 267-290.

Ingham, B. 2003. The Wages of Spin. London: John Murray.

Johnson, K. 2008. Quantitative Methods in Linguistics. Malden, MA: Wiley-Blackwell.

Koller, V., A. Hardie, P. Rayson & E. Semino. 2008. “Using a semantic annotation tool for the analysis of metaphor in discourse”., 15.

Lakoff, G. 2006. Whose Freedom?: the Battle Over America’s Most Important Idea. New York: Farrar, Straus and Giroux.

Lakoff, G. 2004. Don’t Think of an Elephant! Know Your Values and Frame the Debate : the Essential Guide for Progressives. White River Junction, VT: Chelsea Green Pub.

Lakoff, G. & M. Johnson. 2003. Metaphors We Live By. Chicago: University of Chicago Press.

Lakoff, G. 2002. Moral Politics: How Liberals and Conservatives Think. Chicago: University of Chicago Press.

Lakoff G. & M. Turner. 1989. More Than Cool Reason: a Field Guide to Poetic Metaphor. Chicago: University of Chicago Press.

Lakoff, R. T. 1991. Talking Power: the Politics of Language in Our Lives. New York: Basic Books.

Langacker, R. W. 1987. Foundations of Cognitive Grammar. Stanford: Stanford University Press.

Rayson, P. 2009. Wmatrix: a web-based corpus processing environment. Computing Department, Lancaster University:

Rayson, P. 2008. “From key words to key semantic domains”. International Journal of Corpus Linguistics 13: 519-549.

Rayson, P. 2003. Matrix: A Statistical Method and Software Tool for Linguistic Analysis Through Corpus Comparison. University of Lancaster: Unpublished PhD Thesis.

Richards, P. (ed.) 2004. Tony Blair In His Own Words. London: Politico’s.

Seyd P. & P. Whiteley 2002. New Labour’s Grassroots: the Transformation of the Labour Party Membership. New York: Palgrave Macmillan.

Sweetser, E. 2000. “Blended spaces and performativity”. Cognitive Linguistics 11: 305-333.

Van Dijk T. (ed.) 2007. “The study of discourse: an introduction“, Discourse Studies, Sage Benchmarks in Discourse Studies: xix-xlii. London: Sage.

Haut de page


1  We would like to thank Eve Sweetser for her precious remarks on the first draft of this paper, which was originally written for one of her seminars on Conceptual Metaphor Theory. We would also like to thank the two anonymous reviewers from CogniTextes for their helpful comments on the first version of this paper. All remaining inaccuracies are our responsibilities. Finally, many thanks for Paul Rayson for his help with WMatrix access and use.

2  Here is the text of the original Clause IV, written by Sydney Webb in 1917 : “to secure for the workers by hand or by brain the full fruits of their industry and the most equitable distribution thereof that may be possible upon the basis of the common ownership of the means of production, distribution and exchange” (Seyd & Whiteley 2002: XVI). Here is the text of the new Clause IV: « by the strength of our common endeavour… to create for each of us the means to realise our true potential and for all of us a community in which power, wealth and opportunity are in the hands of the many not the few » (Seyd & Whiteley 2002: XV).

3  “New Labour defined itself by not being Labour. Issues on which the break could be highlighted were actively sought. New Labour is not Labour renewed. It is Labour rejected, Labour renounced. New Labour is a negative. New Labour is, and is meant to be, Not Labour.” (Roy Hattersley in Seyd & Whiteley 2002)

4  Fairclough’s New Labour New Language? (2000), is a notable exception. Based on a significant corpus of texts, it provides the reader with some quantitative analysis; the latter is however not detailed enough to vouch for statistical significance.

5  Word frequencies as computed by WMatrix; see below & Rayson 2009.

6  UCREL Semantic Analysis System.

7  See for the entire tagset.

“The semantic tags are composed of:
1. an upper case letter indicating general discourse field.
2. a digit indicating a first subdivision of the field.
3. (optionally) a decimal point followed by a further digit to indicate a finer subdivision.
4. (optionally) one or more ‘pluses’ or ‘minuses’ to indicate a positive or negative position on a semantic scale.
5. (optionally) a slash followed by a second tag to indicate clear double membership of categories.
6. (optionally) a left square bracket followed by ‘i’ to indicate a semantic template (multi-word unit)” (Archer et al. 2002: 1)

8  As a consequence, some of the keyword analyses in this article display different variants of the same lemma.

9  For these last two reasons, we have chosen to use small caps to refer to these concepts in our analyses, as opposed to actual words which appear in italics.

10  See Rayson (2003) for the detail of the calculation of G2.

11  A code yielding p-values for each LL score will soon be added to WMatrix, thus allowing for more precision in the determination of test significance (Rayson, personal communication).

12  Throughout this article, a couple of representative examples from the three manifestos will be provided for each argument, in addition to statistical output.

13  While Google is not a reliable source for corpus analysis, it does nonetheless provide interesting results as far as this combination of metaphors is concerned. A search for "the more the better" gets 1,380,000 hits, while a search for "the more the worse" only gets 25,900 and "the less the better" 29,500. Comparable differences occur with searches for "the more * the better" (3,270,000 hits) vs. "the more * the worse" (128,000 hits) and "the less * the better" (546,000 hits).

14  The forward-back dichotomy will be discussed in more detail in the third part of this article.

15  See Coulson & Oakley (2000, 2006), Fauconnier & Turner (2002) and Grady et al. (1999) for a detailed presentation of Blending Theory and its diagrammatic conventions.

16  While the blend is entitled “Tories are Zombies”, WE admit to using a very liberal definition of the zombie as such; Figure 6 clearly shows that Input Space 1 is already a blend of many different horror movie characters, conflating for example zombies and body snatchers. The important element here is not whether the poster gives an accurate portrayal of zombie attributes (whether a zombie is evil or not for example), but that in the input space the image of the zombie on its own carries the main elements that are characteristic of horror movie characters in general.

17  See similar mappings for the analysis of “My karma ran over my dogma” in Coulson & Oakley (2000).

18  In most of the examples, time is actually not being directly metaphorized. Most occurrences cited are actually entailments of the two main metaphors at play here, namely moving time and moving ego. This does not affect our argumentation.

19  A car is not the only vehicle to have a reverse gear and a driving seat, but it is clearly the first vehicle that comes to mind when these terms are mentioned.

20  Cienki (2005), who provides one of the first empirical tests of the theory, shows for instance that entailments of the “world-view” models outnumber by far the verbal instances of the conceptual metaphors identified by Lakoff.

21  The past has already happened whereas the future hasn’t – thus there are concrete elements to show for the past, which cannot be changed, whereas the future is only a hypothesis in the process of materialising itself.

22  This article extends Evans & Tyler’s analysis of the semantics of English prepositions (2003) to the adverb under scrutiny. Their method is relevant to our case because forward is an adverb with a spatial primary sense and metaphorical extensions.

23  Most of the emergent structure of the blend has already been analysed throughout this paper (especially as far as new Labour’s opponents are concerned) – thus we will not come back to it here.

24  Comment by “MeandYou” on August 29 2008 to a Guardian article entitled: “In praise of… American rhetoric”. Last accessed at [18/11/2009]

25  “Israeli Candidate Borrows a (Web) Page From Obama”, by Ethan Bronner and Noam Cohen, The New York Times, November 15, 2008.

Haut de page

Pour citer cet article

Référence électronique

Emilie L’Hôte & Maarten Lemmens, « Reframing treason: metaphors of change and progress in new Labour discourse », CogniTextes [En ligne], Volume 3 | 2009, mis en ligne le 15 décembre 2009, Consulté le 27 mars 2017. URL :

Haut de page


Emilie L’Hôte

Université Lille 3 & CNRS UMR 8163, Domaine Universitaire du Pont de Bois, BP 60149, F-59653 Villeneuve d'Ascq, France

Maarten Lemmens

Université Lille 3 & CNRS UMR 8163, Domaine Universitaire du Pont de Bois, BP 60149, F-59653 Villeneuve d'Ascq, France

Articles du même auteur

Haut de page