! " # ! $ %&' ( )* + , & The UK’s Faculty of Public Health defines public health as “[t]he science and art of preventing disease, prolonging life and promoting health through organised efforts of society.” (“What is public health? ”). Does this mean that in order to prevent disease and prolong life, society (and, in particular, the state) has a duty to restrict personal choices when it comes to potentially ‘risky’ or ‘unhealthy’ lifestyles? Or does this impair civil liberty and the freedom of the individual to choose for himor herself? Nowadays, government guidelines and recommendations are omnipresent: everyday consumer goods carry labels warning of potential risks and estimating their general suitability for a ‘healthy lifestyle’, and advertisements on television and radio promote ‘healthy choices’ from quitting cigarettes to drinking responsibly to losing extra pounds. Many of these endeavours might be sensible from a medical point of view. However, in the public’s point of view, there is a thin line between being helped and being told what to do. It was Margaret Thatcher who first coined the term Nanny State in 1979 (Cottam 2005: 1593), and it has since been invoked with suspicion whenever government policies seemed to be too intrusive on people’s private lives, dictating what they ought to eat and drink and how they ought to behave. Calman (2009: e9) concludes his summary of the Nuffield Council on Bioethics’ examination of ethical issues in public health with the statement that “the state has a duty to look after the health of everyone, and sometimes that means guiding or restricting people’s choice.” Restriction, and even guidance, of choice is an issue in the public’s perception of these legislations and cam- $$$ - $ ' & $ + & $) ! " # $% # paigns, especially if, as Cottam (2005: 1593) points out, the targets are lifestyle choices. At the same time, the discourse about preventive health and risk has changed in a wider context. Health has become conceptualised not so much as the absence of illness, but through its relation to the risk of potential ill-health. As Skolbekken (2008: 18) puts it, “nobody is perfectly healthy through [surveillance medicine’s] gaze. We are all potentially sick or at risk of developing a disease and eventually dying.” Lifestyle choices are a dominant feature of this discourse. As Petersen and Wilkinson (2008: 7) argue, “[m]any practices and interactions that were once seen as bringing pleasure have become the source of worry”, not just for governments, but for the population at large. A greater emphasis on preventive medicine and potential risks has led to a reappraisal of the (potential) patient’s role: Ogden (1995) outlines the progression from a passive through an interactive to an intra-active identity in psychological theory - once people are conceptualised as being ‘in control’ of their own health, “[t]he risk to health comes from the individual’s presence or absence of self control” (413). The patient, therefore, is no longer simply expected to passively undergo treatment, but to actively ‘work’ towards health. On a larger scale, everyone is expected to take active measures to assure that his/ her health is as good as it possibly can be. Of course, in the context of the increasing preoccupation with risk, this also means avoiding even potentially ‘unhealthy’ factors. This ideological interconnectedness of the concept of health with risk (and its avoidance), in combination with an increasing expectation of active risk management leads Petersen and Wilkinson (2008: 6) to state that “the self-management of risk has become an imperative of citizenship.” The argument is already framed in political terms. However, the authority of the state is especially precarious in matters of personal lifestyle choice, and authoritarian rules and regulations on the matter are connected with the negatively connotated ‘Nanny State’. The question of how public health campaigns try to avoid this stigma in texts focusing on the concept of risk prevention therefore promises to be interesting. , . / +! This paper aims at analysing a number of texts taken from the websites of governmental public health organisations from the UK and the Republic of Ireland (NHS England, NHS Wales, NHS Scotland, Health and Social Care in Northern Ireland and the Irish Health Service Executive). All of these texts offer advice and guidelines to users, and all of them focus on three main lifestyle choices: food, alcohol and smoking. Usually, physical exercise and sexual and/ or mental health are the other main topics in the & ! ' ( lifestyle section of these websites. However, in order to keep the corpus relatively homogenous, only the three sections that are to do with the consumption of goods have been taken into account. The corpora were manually tagged to exclude repeated text from links (such as back to top or for more information see...) from the analysis. The following table shows the numbers of types and tokens in the three sub-corpora. Corpus Types Tokens Alcohol 3,559 31,909 Food 3,809 48,686 Smoking 3,030 31,226 Total 6,834 111,821 % ) ! *% '+ % + , The standardised type/ token ratio for the entire corpus is 38.4, far below the standardised TTR of the British National Corpus at 43.5. This indicates a level of sophistication that is below that of the average (mostly written) British English text, and indeed the texts collected tend to favour a more colloquial language and simpler structure than texts on the same subject geared towards a medical audience. Both Critical Discourse Analysis and Corpus Analysis are employed in order to investigate how the role of the advisee is constructed, and whether this reaffirms the theory of risk and personal responsibility outlined above. Corpus linguistics, rather than being a field of linguistics per se, can rather be termed a methodology. As McEnery and Wilson (2004: 2) point out, “[a] corpus-based approach can be taken to many aspects of linguistic enquiry.” The aim of corpus linguistics is the description of natural language by means of the computerised analysis of so-called corpora. A corpus, according to Teubert and ermáková (2007: 41), is made up of “a (suitable) sample of the discourse” that is to be investigated, since the entire discourse of a speech community would of course be too large to process. Naturally, since most corpora are finite, there is a certain amount of controversy about what constitutes a representative corpus (cf. McEnery and Wilson 2004: 29-31). The software used to analyse the compiled corpora for this paper was Mike Scott’s Wordsmith 5.0. Critical Discourse Analysis, according to Mesthrie et al. (2001: 320), acknowledges the interrelation of ideology, power and language, and attempts to analyse texts accordingly. Its main proponent is Norman Fairclough. In Language and Power, Fairclough (1989: 1-2) draws attention to the way language perpetuates social power inequalities through linguistic conventions of which language users are usually not consciously aware. Fairclough uses the concept of discourse, which he defines as “language as social practice determined by social structures” (1989: 17). He establishes a three layered model of discourse, with the ‘text’ at its centre, surrounded by ‘discursive practice’, which is in turn embedded in ‘social practice’ (Fairclough 1989: 25). Discourse analysis is ‘critical’ because it is supposed to “show up connections which may be hidden from people - such as the connections between language, power and ideology” (Fairclough 1989: 5). Due to the nature of the texts and the public perception of unwelcome interference with personal choice on the one hand, and a healthy lifestyle as one minimising as many risks as possible on the other, certain features are expected to occur: in particular, due to general suspicion concerning the ‘Nanny State’, they are presumably not overtly prescriptive (you must), instead appealing to common sense notions of health and wellbeing. Furthermore, direct references to institutions are expected to be nonovert, as this would reinforce their textual power rather than the reader’s. From a simple reading of the texts used in the corpus, the impression is that, while the word we occurs relatively frequently, it is usually used as a generic expression of common sense (we all) rather than referring to the institution giving advice. Preliminary analysis has identified the most salient semantic fields among the two hundred most frequent words in the corpus. It is to be expected that in a corpus that is primarily preoccupied with lifestyle choices connected to nutrition, smoking and drinking words like food, smoking and drinking appear very frequently. Likewise, texts representing public health organisations and focusing on preventive health can be expected to talk at length about health or risks. However, there were several semantic categories that were surprising, and telling in how they were used in the discourse. These were chosen for further analysis. The following table illustrates the semantic categories devised for the first two hundred words of the corpus, with the token numbers in brackets. Of these, ‘Choice’, ‘Trying’, ‘Reducing, Quitting, Increasing’ and ‘Help and Support’ were analysed in greater detail. Semantic categories Action make (309), use (191), find (146), add (86), found (70), makes (70), using (70), Alcohol and Cigarettes alcohol (653), smoking (642), drink (482), drinking (479), smoke (257), drinks (176), nicotine (137), cigarettes (97), cigarette (90), drunk (72) The Body body (223), blood (138), heart (138), liver (100), skin (59), physical (54) Choice choose (111), choice (53) Evaluation good (239), likely (131), just (140), important (117), better (107), different (97), well (96), instead (92), small (86), best (81), young (69), same (67), sure (65), & ! ' ( especially (63), right (59), easy (58), heavy (58), usually (56), available (55), long (55), fresh (54), added (52), balanced (52) Health, Ill-health and Risk health (268), healthy (250), risk (230), disease (123), cause (115), problems (112), symptoms (95), cancer (92), effects (91), energy (89), pregnant (88), life (86), pregnancy (83), healthier (67), problem (67), damage (65) Food fat (437), eat (330), food (330), foods (300), salt (275), weight (273), fruit (242), fish (233), eating (198), meat (157), diet (153), milk (145), calories (127), sugar (116), water (109), breakfast (107), vegetables (105), saturated (103), bread (101), vitamins (90), eggs (83), rice (82), beans (75), juice (75), vitamin (73), cheese (70), starchy (69), meal (67), oily (67), cooking (63), meals (62), shellfish (62), cooked (61), sugars (61), pasta (59), veg (59), nutrients (57), potatoes (56) Help and Support help (347), support (143), advice (113), tips (78), helps (53) Measurements high (239), units (199), amount (142), low (105), per (91), levels (90), lower (80), lots (68), regularly (65), portion (64), portions (62), half (61), plenty (61), 100g (59) Perceptions and Emotions need (252), feel (167), think (151), want (144), know (143), like (132), remember (118), see (112), look (93), withdrawal (74) People people (351), children (168), women (149), baby (138), smokers (97), men (91), child (88), GP (81), family (64), friends (54), babies (53) Reducing, Quitting, Increasing stop (210), quit (166), keep (127), don’t (124), avoid (109), start (102), reduce (96), quitting (83), lose (70), stopping (67), increase (61) Time day (277), time (190), week (103), years (84), times (64), year (64), days (58) Trying try (257), trying (63) Movement and Transfer give (219), go (168), take (120), giving (88), going (64), put (63) % ) ( * # , $ ! , ( The official website of the NHS England is not simply called “NHS England.” Instead, its header reads “NHS choices. Your health, your choices.” The concept of being able to choose for oneself is important throughout the texts. The verb form choose itself occurs 111 times in the corpus, the corresponding noun choice/ s 77 times, and option/ s 47 times. These occur significantly more often in the food corpus, where the choice is mostly between regular and ‘healthier’ options, as in Try to choose wholegrain varieties whenever you can. In the alcohol corpus, the verb usually occurs in the phrase choose to drink, while the noun only occurs three times: Concordance sample of choice* ot drink is a much healthier choice . The NHS recommends that: * Men keep count and make informed choices when you're drinking. Bottle (330ml) more than you that's their choice * Don't mix alcohol with drugs of an The first of these repeats the pattern from the food corpus, where there is a healthy and therefore, by implication, an unhealthy choice that has to be made, while the other two stress the importance of personal choice. The latter reappears in the smoking corpus, where the personal choice to smoke or to quit smoking is the most important aspect. Concordance sample of choose* / choice* rch shows that no matter when you choose to give up, be it 18 or 80 years of t them regardless of whether they choose to continue smoking or not. Assure earn: * Remind yourself that your choice to smoke or not to smoke is still the The frequent use of the word choose, often occurring in speech acts such as advice or suggestions, usually represented by imperatives, and very often with direct reference to the reader (you/ your) in close proximity, stresses the concept of personal action. It is the reader who must ultimately choose what to do about his/ her situation, and whether to act according to the advice given on the homepages or not. Of course, the semantic implication of choose and choice also touches on the issue of personal freedom, the value that is seen under threat by the ‘Nanny State’ outlined in the Introduction. When advice is presented as a choice, it ceases to sound restrictive; after all, a choice implies always at least two possible ways of action (even though it might be objected that there is a contradiction between using a directive speech act in connection with choice). Presenting a healthy lifestyle as an individual choice avoids an authoritarian “thou shalt not” attitude, which people might be wary of. However, what constitutes a healthy lifestyle is still defined by the institutions giving advice, and whether the choice is actually a sheer matter of free will is debatable. In one of the videos on the NHS website, a young woman talks about weight loss in terms of choice, but the example she gives is interesting: It’s [i.e. losing weight is] so much of a lifestyle choice now. You know, do I wanna be really unhappy, or do I wanna be en- & ! ' ( joying life? (“Weight loss tips” [online]). The commonsensical assumption revealed by this statement is significant: real enjoyment comes from being slim and active and leading what the NHS promotes as a healthy life, while any deviance will result in unhappiness. Emotive language like this does raise the question whether people who do not conform to this lifestyle actually feel unhappy, and if so, whether this is because public discourse constructs them as unhappy. If one is told that refusing to choose the ‘right thing’ will make one miserable, does this still count as a free choice? , ! + Another semantic field that occurs with striking regularity is that connected to ‘trying’ things. Interestingly, this semantic field reappears throughout all three sub-corpora with roughly equal frequency. The word try appears 257 times in the corpus, and trying appears 63 times, while the past tense is significantly less present: tried only appears 16 times, so the focus is obviously on the ongoing process, rather than the results of it. Its synonym attempt and its grammatical forms, however, only appear fifteen times, of which four refer to attempted suicide, and only eleven are used in phrases like Concordance sample of attempt* most smokers several attempts to quit completely. The good n elp you in your quit attempt if you’re prepared for the chal prepare you for this attempt and help you avoid temptation. metimes stem from an attempt to cope with an underlying prob The use of try rather than the slightly more formal attempt ties in with the less sophisticated nature of the source texts, which is already detectable in the standardised type-token ration. It is, however, also noticeable in terms of its grammatical usage: in the relatively informal ‘language of advice’ the texts in the corpus are employing, it sounds far more natural to use the word try as an imperative than the word attempt. The word try is particularly interesting in this context: it is only used as a verb, and apart from the phrases you could try and you should try and the infinitive use of to try, it is always used as an imperative. This gives a total of 213 instances of an imperative need to try something. The five clusters that appear most frequently are (in order of their frequency) try to eat, you could try, try not to, try to choose and try to avoid. Of these, you could try seems to be a suggestion, while the other forms seem to realize a more straightforwardly directive speech act. Fairclough (1989: 55) draws attention to the hidden power in directives and hints, a strategy that is sometimes more effective because it draws less attention to the power imbalance that is expressed. You could try (and also the related phrase you should try, see above) quite obviously falls within this category. However, I would argue that even when it is used in an overt imperative structure, such as try to choose the verb try retains enough semantic ambiguity to avoid sounding coercive. In a construction like try to choose wholegrain varieties, try to choose the healthier options or try to choose options that are low in fat, one can even find three words that soften the imperative meaning in a single phrase: try, choose, and options/ varieties. All of these leave room for the ‘free will’ of the advisee: choosing between different options or varieties has already been discussed above. Telling readers to try to eat certain types of food does not sound nearly as authoritative as simply telling them to eat certain types of food. Focusing on trying rather than achieving includes the possibility of failure, thereby making it less threatening - one could even argue that this gives the reader an easy way out if their attempt is not successful, while at the same time urging them to act. It also casts the institutions in the role of advice providers rather than commanders - there is a certain implication of understanding in this insistence on trying. This observation ties in with the general avoidance of ‘restrictive’ directives in the corpus: the phrase you should appears 57 times in the corpus, whereas the much more rigid you must only occurs three times, and at one of these occurrences, is used in an subordinate clause, thus not contributing to a directive illocution (reduce the risk of destructive ‘all or nothing’ thinking that says you must either stick to an extremely rigid diet or no diet at all). Moreover, the empowering phrase you can occurs a staggering 304 times, making it clear that the focus of the texts is on an enabling rather than a prescriptive discourse. Again, this underscores the active role of the advisee, and the concept of free will and the freedom to choose attached to it. , , & + 0& + A peculiar aspect that became obvious during the preliminary analysis is the use of verbs connected, in a wider sense, to increase and decrease. In this respect, there are clear connections of certain verbs with certain subcorpora. While the food corpus, as mentioned above, is preoccupied with choosing a different (by implication a healthier) option, and therefore with a change in diet, the smoking and alcohol corpora seem to reflect the acceptability of the habit. While the consumption of alcohol seems to be acceptable within limits, smoking is seen as a habit that should be abandoned altogether. The word stop, occurring 210 times in the corpus, is predominantly found in the texts that advise against smoking. Stop smoking is by far the most frequent phrase in this context, occurring 72 times, and thereby far & ! ' ( more often than stop drinking, which only occurs 8 times, with one of the occurrences being an advice to never stop drinking suddenly. Noticeably, in the food and alcohol corpus, the word tends not to be used in connection with an action on the part of the readers; they are not advised to stop doing something, but rather informed about possible outcomes, as in the following examples: Concordance sample of stop* alcohol quota out for longer. It'll stop you getting dehydrated too and les e body, causing changes in some and stopping others from working properly. t eat properly and too much alcohol stops the body absorbing the nutrients ohol disturbs our brain rhythms and stops us getting enough of the deep sle ing sex. And of course, it can also stop guys from getting and keeping an e minutes to help you lose weight and stop weight going back on n ater, or other fluids, every day to stop us getting dehydrated. When the we re helps to keep bowels healthy and stop constipation. And this means we ar With the word quit, which is more closely linked idiomatically to the phrase to quit smoking, this is even more noticeable. 261 occurrences, all but one in the smoking corpus, and without exception linked to the concept of quitting tobacco, drive home the importance attached to giving up cigarettes for good. Smoking seems to be the only topic considered in this paper where a governmental insistence to cease and desist is considered acceptable, and even an imperative like Quit smoking for the benefit of you and your family does not appear to be too harsh. When the concept of reducing the intake of certain substances or ‘cutting down’ on them is considered, the relation between the alcohol and the smoking corpus is reversed. While the reader is advised not to drastically stop a problematic alcohol intake, but rather cut down on it, cutting down on cigarettes is discouraged in favour of giving them up completely. The phrase appears in the smoking corpus in sentences like Can’t I just cut down rather than give up [smoking]? (a question which is answered in the negative), or the warning that when you cut down you tend to take more and deeper puffs on each cigarette. In the alcohol corpus, however, where the phrase appears most commonly, cutting down is encouraged. Generally, there is a certain preoccupation with amounts and measurements in the alcohol corpus, while the smoking corpus focuses on indicators of time (days, months, years). The texts in the smoking corpus go to great lengths to describe the physical effects of quitting cigarettes after a certain amount of time. There is no recommended cigarette intake over the course of a week, while most of the words connected to time in the alcohol corpus refer to the recommended limits of one’s daily or weekly alcohol intake. A word closely connected to the concept of re- stricting one’s alcohol intake to what is perceived as a reasonable level is the frequently occurring unit. This measurement occurs 236 times in the corpus, all but five times in the alcohol sub-corpus, and every single time it refers to a so-called standard unit of alcohol, defined as 8g of pure alcohol, which is used to measure the alcohol intake for men and women that is still deemed medically safe (cf. Jackson and Beaglehole 1995: 716). The focus of the text here is first to define a unit in generally understandable terms, and then give clear guiding lines as to how many units are still deemed within a ‘healthy’ range, e.g. Concordance sample of unit* -strength bitter contains just over two units , while a glass of wine can contai alf units. A double can be nearly three units . You can't count one drink as one y should not drink more than one to two units of alcohol once or twice a week an It seems that, unlike smoking, which is explicitly condemned, alcohol is still so widely accepted socially that the language used in connection with it refrains from any outright ban and rather tends towards reducing its use as part of the desired ‘healthy lifestyle’. ,1 * "&** Despite its overall focus on empowerment and the agency of the reader, the concept of help is omnipresent in the corpus. The word help ranks seventh in a list of content words of the entire corpus, before words like food, health or risk, and even higher if grammatical variants such as helps and helping are included. All in all, there are 416 occurrences of the word help/ s, 143 occurrences of support, but only eighteen instance where the word aid(s) is used. Interestingly, the more sophisticated aid is only used in the smoking corpus, to refer to cessation aids, i.e. chemical nicotine substitutes. The frequent offers of support seem, at first glance, to cast the advisee in a more passive role than originally expected. The implication seems to be that they are struggling with some kind of deficiency that requires help, whether it be the ‘wrong’ diet, an addiction to nicotine, or the overconsumption of alcohol. At the same time, the omnipresence of offers to help reinforces the ‘benevolent’ nature of the discourse: rather than presenting people with a set of behavioural rules in order to make them change their lifestyles, they are repeatedly told that doing certain things will help them. Interestingly, it is hardly ever people who are the source of help. A personal helper is only mentioned in twenty-four instances, and most of the times, it is people from the personal environment of the advisees who act in this role - friends, family, the family doctor, or even they themselves, but not the institution giving the advice in the first place. The only institutional entities to appear in this context are local voluntary services, & ! ' ( smoking cessation advisors, and regular support groups all mentioned once. Ten times, the corpus advises to seek help from your GP or your doctor, four times it is friends and family that are identified as helpers, twice the phrase help yourself appears, and in the remaining five instances, it is the reader him/ herself who can help other people of their acquaintance (to give up smoking in all instances). Once again, this avoids giving obvious discursive power to political institutions like the NHS. Help, when it comes from a human source, is limited to the personal environment, or at the most to local and non-governmental organisations that are not tinged with the suggestion of influence from the state. The vast majority of help, however, is not portrayed as stemming from human beings at all - it is abstract concepts, ways of behaviour, advice and information, or drug treatments that promise to help the reader. This is a prominent feature of all three sub-corpora, exemplified by phrases like the following ones: Concordance sample of help* And eating breakfast could actually help you control your weight. There i the oil as you can and it will also help to put the egg on some kitchen pape take your mind off your drinks and help you to slow down. Step 7 Have alc ating before you start drinking can help limit how quickly alcohol gets into otine replacement therapy (NRT) can help people break this physically addict juice There are treatments that can help reduce the intensity of these cravi g smoking. When you give up, it can help to know that the first few days are utting down on tea and coffee could help to improve iron levels in the body. This way of insisting on the helpful quality of behaviours that are to be encouraged not only avoids giving more direct commands that might sound too authoritarian. It also implies that the outcome that one is helped to achieve is ultimately desirable. If eating breakfast is advertised as help[ing] you control your weight, controlling one’s weight according to the standards outlined throughout the rest of the text is presented as a positive and favourable outcome, without having to resort to more restrictive rhetoric. 1, ( & Throughout the corpus, advisees are constructed as active participants rather than passive recipients of advice, and the concept of free will is strongly emphasised in connection with their actions. The texts steer clear of phrases and words that might sound too restrictive and thereby cast the institution giving advice as dictatorial or reinforce the stereotype of a state that aims at controlling its citizens. This is accomplished by a strong focus on choices, implicitly condoning a pluralist attitude that allows for alternatives, and a preference for imperatives that sound less threatening, e.g. try doing instead of do. The concept of help, which is omnipresent in the text, is mainly linked to the private sphere if the helpers are human, but far more often used to refer to modes of behaviour or treatments. The governments or the public health institutions from whom the advice stems, however, hardly ever appear as helpers, arguably to avoid drawing attention to them as more powerful than the reader. The discourse also seems to echo cultural norms, such as the avoidance of a complete condemnation of so powerful a cultural signifier as alcohol. Yet despite the empowerment of the advisee and the strong focus on free will and choice, commonsensical notions of ‘good’ and ‘bad’ decisions are reinforced throughout the corpus, making the notion of an actual ‘free choice’ questionable. All of this evidence points to the conclusion that the ‘Nanny State’ discourse has indeed caused a shift towards ‘soft’ advice, with a focus on personal freedom and the freedom to choose, resulting in the avoidance of authoritarian commands. ' + * ! Calman, Kenneth (2009) “Beyond the ‘nanny state’: Stewardship and public health.” Public Health 123: e6-e10 (sic). Cottam, Rachel (2005). “Is public health coercive health? .” The Lancet. 366.9497: 1592-1594 Fairclough, Norman (1989/ 1996). Language and Power. London: Longman Jackson, Rodney & Robert Beaglehole (1995). “Alcohol consumption guidelines: Relative safety vs. absolute risks and benefits.” The Lancet. 346.8977: 716. McEnery, Tony & Andrew Wilson (2001). Corpus Linguistics. 2 nd ed. Edinburgh: Edinburgh UP. Mesthrie, Rajend, Joan Swann, Andrea Deumert & William L. Leap (2001). Introducing Sociolinguistics. Edinburgh: Edinburgh University Press. Ogden, Jane (1995). “Psychosocial theory and the creation of the risky self.” Social Science & Medicine. 40: 3. 409-415. Petersen, Alan & Iain Wilkinson (2008). “Health, risk and vulnerability: an introduction.” In: Alan Petersen & Iain Wilkinson (eds.). Health, Risk and Vulnerability. London: Routledge. 1-15. Scott, Mike (1996). WordSmith Tools 5.0. Oxford: OUP. Skolbekken, John-Arne (2008). “Unlimited medicalization? Risk and the pathologization of normality.” In: Alan Petersen & Iain Wilkinson (eds.). Health, Risk and Vulnerability. London: Routledge. 16-29. Teubert, Wolfgang and Anna ermáková (2007). Corpus Linguistics. A Short Introduction. London: Continuum. “Weight loss tips” NHS choices. http: / / www.nhs.uk/ video/ pages/ medialibrary.aspx? Id ={B56BA25D-B993-4206-8A824F18F30F313F}&Uri=video/ 2007/ sept/ Pages/ Weightloss.aspx. (November 2009) “What is public health? ” (n.d.) The Faculty of Public Health. www.fphm.org.uk/ about_faculty/ what_public_health/ default.asp. (November 2009) & ! ' ( " & Health and Social Care in Northern Ireland. http: / / www.healthandcareni.co.uk/ (November 2009). HSE.ie. Health Service Executive Website. http: / / www.hse.ie/ eng. (November 2009). NHS Choices. Your Health, Your Choices. http: / / www.nhs.co.uk (November 2009). NHS Scotland. Scotland’s Health on the Web. http: / / www.show.scot.nhs.uk (November 2009). NHS Wales. Health of Wales Information Service. http: / / www.wales.nhs.uk (November 2009). $ - # . /