Skip to navigation – Site map

Patterns and Variation in the Weather Forecast: Can Prosodic Features be Predicted Too?

Susan Moore Mauroux

Abstracts

The aim of this article is to explore a specific type of media programme: the weather forecast. The analysis is based on a corpus of weather forecasts for the UK, mainly from the Met Office, with five different speakers. The weather forecast is considered as a specific oral discourse type which functions within a fairly set framework involving a number of recurrent features, including a well-defined lexical field and recognizable prosodic patterns. This analysis will try to determine what these features are and how they interact. The study reveals how lexical and prosodic cues are used to connect with the target audience, enabling the forecaster to structure, highlight and even comment on information. The study aims to bring to light what our expectations are as regards the weather forecast, and how far these expectations affect our perception and even our understanding of the message.

Top of page

Full text

Introduction

1The weather forecast is a very typical programme on both radio and TV, and can also be accessed via the internet. The BBC celebrated 60 years of broadcasting the weather forecast live in 2014, and the Met Office – which is the source for my corpus – has been forecasting for over 150 years. Yet, even though it is clearly aimed at the general public, it is also a highly specialized field with a certain number of specific characteristics. It is these characteristics which this article will attempt to define within this multimodal context. I will look at prosodic features as well as other significant characteristics of the weather forecast; and I shall try to see what challenges it presents in terms of the perception of its message.

1 The Corpus

2The corpus focuses exclusively on the UK, and mainly on the Met Office weather forecast updates available on the internet,1 but it also includes one forecast by Michael Fish from 19792 so as to appreciate the changes over the past few decades. Beyond the fact that these forecasts necessarily include a certain amount of variation in the weather itself, they also imply clear focus on the informational aspect associated with visual aids since the target audience has freely chosen to access a website.3 A complementary corpus was also considered and includes other recordings of weather forecasts by non professionals,4 and recordings of some of the professional forecasters outside the context of the forecast, but on the subject of the weather.5 The recordings were scripted and then analysed from visual, lexical6 and prosodic perspectives, using the acoustic analysis software PRAAT (Boersma and Weenink, 2014).

2 A particular style of discourse

3When tuning in to the weather forecast, it is quite easy to identify the type of programme, which leads me to consider it as a “genre”. Indeed, the definition given in the Oxford Advanced Learners Dictionary, 7th edition, 2005, states that a genre is: “a particular type or style […] that you can recognize because of its special features”. It does, however, specify “literature, art, film or music”, so the weather forecast may not truly be considered as such, but it does nevertheless possess at least three very recognizable features.

  • 7 This phrase was used by Hélène Chuquet, University of Poitiers, as an object line in her email enqu (...)

4First, the forecast corresponds to a pre-established discourse situation involving a technical presentation with a generally neutral standpoint. It will also make use of a specific lexical field both to refer to the weather and to structure the discourse. Thirdly, it has recognizable prosodic patterns. These three characteristics may be exemplified in the phrase: “sunshine and scattered showers”.7 Indeed, the phrase has a predictable prosodic pattern as follows: /sunshine / and scattered showers/ with a low fall on the first tone unit and a higher fall on the second, as an echo of the first; it includes specific weather terms and has a fairly high information content; and it is an expression which is frequently encountered in the weather forecast. I will be looking more closely at both lexical and prosodic features of the weather forecast later on, but let us first look at the general characteristics of this particular type of media.

5The weather forecast constitutes a specific communication situation, according to the criteria proposed by Sandré (2013) when defining discourse types. She in fact quotes the weather forecast as an example of what Wichmann calls “goal-oriented” discourse (Wichmann 2000): « Certains genres ont un but très précis (le bulletin météo par exemple, qui a obligation de donner des indications météorologiques sur les jours à venir) » (Sandré 2013: 30). Sandré also identifies the time-space framework and the participatory framework as relevant for the analysis, where Wichmann’s classification refers to oppositions between public/private and monologue/dialogue. Using these criteria, I propose to define the weather forecast discourse situation which is under study as follows: a public goal-oriented situation involving regular studio recordings whose aim is to give information regarding the weather for the next few hours or days; the information is accompanied by visual representations, and given by just one person without any apparent interaction.

6As Silber-Varod and Kessous (2008) underline in their case study based on weather forecasts in Hebrew: “The goal of the speaker is to convey a maximum amount of information in a minimum amount of time and in the most comprehensible way possible.” Beyond the time constraint, the discourse situation represents a specific challenge as the speaker is a professional meteorologist who possesses technical information about the subject enabling him/her to use technical terms and representations. The listener, however, represents a target audience which has a priori no shared specialist knowledge, so there is a considerable competence gap between the speaker and the listener, whether the listener is a native speaker or not. What does the speaker do to bridge this gap and enable the listener to understand all this information?

2.1 Visual cues

7One of the specificities of the weather forecast is their use of graphic representations, as we can see in Figure 1. The first section (green) recapitulates how they use immediately understandable charts or maps which demonstrate visually the situation the meteorologist is describing. The second section (orange) corresponds to more technical forms which need to be accompanied by a certain amount of comment and explanation to be understood.

Figure 1: graphic representations

Figure 1: graphic representations

8The table shows how much the weather forecast has evolved since 1979 as Michael Fish was simply unable to use many of the static and immediately understandable representations which are available today; of course, today’s presenters make extensive use of them, thereby communicating well with the uninitiated target audience. But if Michael Fish had no other option but to use more technical representations, today’s meteorologists seem to vary considerably in what technical representations they use, and how much they use them. Yet, whatever style the meteorologist decides to adopt, these visual aids are always accompanied with explanations and he/she uses gesture (pointing to the appropriate part of the country or meteorological tendency), again helping the audience to follow the explanations. Moreover, he/she also makes frequent use of the verb “see” (particularly “we see”), which shows a definite attempt to communicate with the audience, even identifying with the listeners. This will be further studied in the section on the use of lexical choices, but it does already raise the question of how the speaker reaches the target audience in this particular oral discourse situation.

2.2 A general pattern

9Let us now consider what the general characteristics of the weather forecast are, as regards the “text” itself. According to Silber-Varod and Kessous (2008), “it is planned speech where a skilled reader reads out a previously constructed text.” This was indeed my own expectation when I set out to research this subject, because we tend to associate the weather forecast with the news which is clearly scripted. Yet I was surprised to find that in the corpus under study this was not the case. Crystal (1995: 296), who also lays emphasis on the time constraint, considers the weather forecast as unscripted and spontaneous discourse: “the visual material is prepared in advance, but during the broadcast the spoken commentary is spontaneous.” This spontaneity is at least partially reflected in the pace at which they speak.

Figure 2: the corpus, pace for professional weather forecasters

Figure 2: the corpus, pace for professional weather forecasters
  • 8 Crystal speaks of surreptitiously recorded conversations, where the pace reaches “well over 300 syl (...)
  • 9 One of the objectives of the course is reading aloud from a script.
  • 10 Channel 4 news : https://www.youtube.com/watch?v=56crttyLVJQ

10Unprepared speech such as we find in conversation can be very quick. The pace of the weather forecasters in our corpus is quick too (between 201 and 246 wpm), certainly quicker than what we would expect in a radio programme,8 for example. And indeed, I saw that this is one of the major differences between professional forecasters who speak from their notes (they have of course prepared the forecast themselves) and non professionals reading a script. Prince Charles only reads at a rate of 178 wpm, although he is used to public speaking, just not the weather forecast. L2 students9 read between 120 and 170 wpm, just one or two of them around 200 wpm. Likewise, when listening to a TV report on the weather,10 doubtless read from a script, the reporter only speaks at 153 wpm.

  • 11 These performance errors were made by Phoebe Smith in the 19/03/14 and 6/02/14 updates. This contra (...)
  • 12 Compare examples 3 (the final version) and 4 (false start) in the list of examples in the Appendix. (...)

11If quick pace is one characteristic of unscripted discourse; another is that we make performance errors because we are trying to speak without hesitation, as fluidly as possible. For example, we hear “the Scotland” (the forecaster was going to say “the north” and changed her mind in mid-sentence), and “a amber warning”… but the forecaster generally just carries on regardless (there is a time constraint to be respected).11 Of course the sort of mistakes a professional forecaster makes are very different from those made by non-professionals: for example Prince Charles cuts a technical term in the wrong place: “and this weather ah front pushing northward”. Yet, it does happen that a Met Office broadcaster stumbles and decides to begin again, like Phoebe Smith in a January broadcast (15/01/14). Interestingly, the false start is recorded so we can see that the same information is given but formulated with subtle changes12 (but much the same prosody).

2.3 On the lexical front: a varied register

  • 13 The corpus has 1118 word types, for a total of 10 303 word tokens in all. The statistical analysis (...)

12With regard to the discourse type, we expect an informative forecast involving a specialist lexical field, and indeed a study of the forecasts13 reveals some use of certain specialized terms, as shown in the following table.

Figure 3: specialized terms / frequency

  • 14 “The detail” refers systematically to the satellite picture, never to the more common sense.

Specialized terms (frequency)

Specialized compound terms/ MWUs

pressure (22)

low pressure; high pressure; pressure chart; pressure pattern

system(s) (12)

pressure system

front(s) (9)

cold front; warm front; weather front

frontal (6)

frontal system

band (9)

band of rain

isobars (7)

tightly-packed isobars

detail14 (6)

13As we see, the specialist lexicon involves many compound word formations such as “pressure” forming “low” or “high pressure”, “pressure system”.

  • 15 The pairs “south-southern”, “west-western” “north-northern” (without occurrences of “northern Irela (...)

14As the next figure shows, the specific lexical field, with words such as “rain”, “shower, “wind(s)” or “temperature(s)” are also largely represented, as are references to geographical location in the UK together with directional or descriptive adjectives like “south/ southern”, “north/ northern”.15

Figure 4: words referring to weather / frequency (1018 word types)

Word (rank)

N° of tokens

Word (rank)

N° of tokens

Word (rank)

N° of tokens

rain (17)

96

Scotland (53)

40

south (14)

100

showers (32)

57

UK (53)

40

west (19)

87

weather (33)

56

England(54)

39

north (21)

77

temperatures - temperature (51)

41 + 3

Wales (93)

24

east (44)

45

cloud (58)

37

Ireland (105)

21

northern 49)

42

snow (70)

31

western (95)

24

sunshine (71)

31

southern (107)

21

winds – wind (88)

26 + 18

eastern (173)

12

central (239)

7

15Likewise, there are numerous time references (“through the day/ morning/afternoon”) and prepositions are widely used to specify time and place. In fact, “through” is used some 54 times to refer to both time and space, and “across” is the tenth most frequent word in the corpus with over 100 tokens. There is also a predominance of verbs indicating movement, change or influence.

  • 16 See Example 1, Michael Fish. The phrase is also used by Aidan McGivern (26/01/14).
  • 17 Markers of degree include the following (number of tokens): “a bit” (20), “a little bit” (23), “pre (...)
  • 18 In addition to being colloquial, both the adjective « chilly » and the adverbs modifying it seem im (...)
  • 19 Colloquial expressions are often used: “a mixed bag” (Example 2), “when the sun does pop through” ( (...)

16Yet, despite the number of words referring to the specialist field, we in fact have a very variable register in terms of specialization or technicality. Crystal (1995: 385) refers to a mixture of “controlled informality and friendly authority.” Forecasters may speak of isobars, but they will talk about temperatures which struggle (the weather is often viewed as a battlefield where the elements are engaged in a fight), and we also hear certain set expressions like “we’re not out of the woods yet”16 to warn listeners that, despite improving conditions, the weather may continue to be unpleasant. A good example of the use of a clearly oral register is the example of the word “chilly” which is used to refer to cold(ish) temperatures. The word is an informal one, used conversationally, but we would not expect it to be used in a formal context. Moreover, the co-text of “chilly” frequently features adverbs like “fairly”, or “pretty”,17 which also correspond to an oral register;18 and it also appears in very colloquial expressions like “a little bit on the chilly side”.19 These are what Crystal terms “fuzzy expressions” (Crystal 1995: 169), which, he suggests, play a role in efficient communication by relating to what the “ordinary” person understands. Indeed, the target audience does not possess shared specialist knowledge which explains why forecasters choose to use so many expressions belonging to a common, and even informal, register, thereby bridging the competence gap.

  • 20 My thanks to Ramon Marti-Solano, University of Limoges, for his helpful suggestions on collocations (...)
  • 21 If we hear “outbreaks”, this automatically implies “rain”.

17Another characteristic of the weather forecast is widespread use of collocations and set phrases.20 This implies that the listener is often capable of predicting the complete phrase just by hearing part of it, as in the case of “sunshine and scattered showers”. Indeed, “scattered” will almost systematically be found with either “rain” or “showers”; “blustery”, another word from a more informal register, can be found with “day”, “showers” or “wind”. If a native speaker is asked to give word associations, the reply will be almost instantaneous: “outbreaks” + “rain”; “patchy” + “fog” “sunny” + “spells”, “chilly” + “temperatures”. This makes the weather forecast easier for native speakers to understand as they can predict to a great extent, and in the case of certain collocations, the information is even doubled.21 For non-native speakers, however, this clearly constitutes a comprehension barrier because they do not make these associations, and in any case the words are unfamiliar to them. Indeed, McCarthy and O’Dell devote a section of their work on collocations to the weather forecast (McCarthy and O’Dell 2005: 30-31).

2.4 A multi-modal situation: between the speaker and the audience

18Connecting with the target audience is obviously a real issue, and using an informal register is one way of doing this. It is so widespread that one is tempted to speak of interaction between speaker and listener, although this is, of course, impossible in view of the participatory framework. There is, however, a real attempt to relate to the listener in a variety of ways.

  • 22 It is interesting to note that in 1979, Michael Fish uses “you” rather than “we”. His use of modali (...)

19In this multimodal oral situation, gesture is used, although it mostly accompanies what can already be seen on the screen: pointing to a place, or showing a movement. This is often associated with the demonstratives “this” and “that”, and the forecaster clearly identifies with the audience using the pronoun “we”.22 But subjectivity goes beyond using the first person. Forecasters use sensory verbs like “feel” to show how the weather conditions they are predicting will affect the listener (as well as themselves).

  • 23 Other examples of this combination of using an oral register and giving a point of view are: “the b (...)

20Value judgements are also made, as in Example 12: “temperatures are still pretty good for the time of year….”. Adjectives like “disappointing” (Examples 3 and 6), “lucky” (Example 3) or adverbs such as “unfortunately” (Examples 10 and 11) or “thankfully”, (Example 8) all show a definite point of view23 regarding the conditions which are described. This analysis suggests that the way the forecast is expressed reflects the audience’s expectations rather than the neutral informational perspective which might be expected in this discourse situation.

21Finally, forecasters make extensive use of oral discourse markers to link and/or structure their discourse: “so”, “and”, “but”, “well” often occur at the start of an intonational phrase. From a lexical perspective, then, we have a mixture of the formality associated with a specialist context, together with a considerable number of oral markers. There is a certain degree of identification with the listener on the part of the forecaster. The weather forecast is not read from a script; it is communicated in oral mode, which brings us to the study of prosodic features. Beyond the choice of words, how and why does the forecaster use prosody to communicate with the target audience? What patterns can be found?

3 Prosodic patterns

22As regards this corpus, the first pattern we saw is the jingle at the beginning of all the Met office forecasts, signalling the beginning (and end) of each update. When we first hear the tune, we expect the weather forecast update to begin and tune in to receive a certain type of information. In my analysis, I shall study prosodic patterns in relation to this information, to see how they affect our perception of the forecast.

  • 24 Pitch corresponds to the acoustic feature, fundamental frequency (F0), measured in Hz, as seen in t (...)
  • 25 Variation between speakers also needs to be taken into account as different speakers make use of di (...)

23The recordings were first considered from an auditory / perceptual perspective, as well as from a visual perspective, then the audio extractions were viewed using PRAAT so as to appreciate the potential relevance of certain perceptual features, especially pitch24 (pitch height, pitch movement, relative pitch differences, and maintained pitch height).25 Intonation patterns will be analyzed in terms of pitch movement (fall, rise) and the nucleus (Wells 2006; Ashby and Maidmen, 2005). Likewise, I will use the IP (intonational phrase) as the basic unit for my study which aims to explore what prosodic features are used in this type of discourse, particularly in so far as they relate to discourse structure, topic change and highlighting.

  • 26 When making a transcript of the weather forecast, I realized how circular the information content i (...)

24Indeed, weather forecasts as a “genre” present two potential difficulties. First, the information content is by definition repetitive and varied at the same time, because the forecaster refers to the same weather conditions or to the same places at different times or on different days. This can be confusing for any listener.26 Second, as has already been stated at the beginning of the article, forecasters have a very quick pace. The analysis will enable us to determine how far prosody plays a role in helping us follow this very quick pace, and what its role is as regards overall discourse structure.

  • 27 He lays considerable emphasis on the modal “could” in “where it could lead to difficult conditions (...)

25Quick pace is often said to present a comprehension difficulty, which implies that slower elocution is easier to follow. However, this is not necessarily true. In the case of Prince Charles reading the weather forecast, for example, although he is considerably slower, he is not easy to understand because the prosodic markers he uses correspond to the type of discourse he is accustomed to producing (public speaking including topics for debate). He therefore uses prosody to emphasise certain words,27 but not to structure the different parts. We therefore lose the prosodic cues we expect in this type of discourse situation. I shall now attempt to define what this prosodic style is.

3.1 Prosodic cues and discourse structure

26The situation implies going from one topic to another (different aspects of the weather, different parts of the country), like presenting the news headlines. New topics can be introduced by an oral connector (and, but…) which is usually produced on a much higher pitch than the preceding words. This process, known as pitch reset (Wichmann 2000), can be seen in Figure 5.

27

Figure 5 - Example 3: Phoebe Smith 15/01 – discourse markers and pitch

Figure 5 - Example 3: Phoebe Smith 15/01 – discourse markers and pitch
  • 28 The pitch contour (F0) is circled in yellow.
  • 29 The same prosodic markers apply to the next topic which is introduced by “if we take a look….”, whi (...)

28The IP begins very high and gradually lowers over the phrase, reflecting expected declination and slowing towards the end.28 This is followed by much higher pitch for the next piece of information (highlighted).29 This is typical of the way information structure is conveyed through prosody, and corresponds to Wichmann’s findings for beginnings and ends (Wichmann 2000). Moreover, in this extract, as well as the very obvious pitch reset from “through the day” to “but”, the connector is also preceded by a pause which contributes to highlighting this topic change. We also see that after “but”, the forecaster narrows the previous focus of the information (“many of us”) down to “some of us”. By producing “but some of us” on the same level, as high maintained pitch, the speaker shows the difference with the previous information, implied in the lexical choice of connector, “but” thereby highlighting the contrast between “many” and “some” both lexically and prosodically.

  • 30 See Herment (2010) and Moore Mauroux (2012) for a more detailed analysis of prosodic prominence wit (...)
  • 31 Prosodic prominence also occurs in indications of scope (“most”, “many”, “some”) or degree (“pretty (...)
  • 32 We frequently find that after marking and highlighting the idea of continuation, the information we (...)

29In the next example (Example 9), we again see how high pitch marks the beginning of an IP which then lowers over the rest of the IP, but here pitch also enables the speaker to highlight information. Throughout the corpus, prosodic prominence30 is used to emphasize significant facts31 (events, locations, tendencies), to contrast them, to show new tendencies and changes, but also continuation. This is the case here with very high pitch on “still”, drawing attention to the fact that the subsequent information is similar to what we already experience.32

Figure 6 - Example 9: Margaret Emerson 12/02/14 - the focusing function

Figure 6 - Example 9: Margaret Emerson 12/02/14 - the focusing function
  • 33 The nucleus of the IP is circled in black (both the pitch contour and word) here and in Figure 7.
  • 34 In this case, “too” is the nucleus of the third IP and so the new additional information is surroun (...)

30Pitch movement associated with the nucleus33 of the IP is also significant as the information corresponding to the nucleus is always prominent to some degree. Moreover, the choice of tone influences our perception of the message. Indeed, in this example, the nuclei for the first two IPs are respectively “breezy” and “showers”. The falling nuclear movement on “breezy” starts high, making it very prominent, whereas the second has only a low fall, implying that this is not as important as the information given previously. It is this type of marker which enables the listener to tune in to the most significant facts. The other word which stands out in terms of pitch is “also”, thereby highlighting additional different information which here contrasts with the conditions given previously. This variability in weather conditions frequently occurs and is reflected prosodically by this type of highlighting.34

31Example 11 is similar to the preceding ones in several ways. First, as we see in the next figure, high pitch is used at the beginning of the IP on “unfortunately” which indicates the change in topic. We also have focus on the words “unsettled”, “further” and “turning”, which is reflected in terms of pitch.

Figure 7 - Example 11: Marco Petagna 18/02/14 – focusing and commenting

Figure 7 - Example 11: Marco Petagna 18/02/14 – focusing and commenting

32For “further”, we have another example of highlighting to show continuation, in this case, of rain; conversely, “turning” indicates a change in trend. And for “unsettled”, there is a falling pitch movement as the word is the nucleus of the IP; indeed, it captures the essential information regarding weather conditions.

  • 35 This is a stylistic feature which is frequently used by some forecasters, including Marco Petagna.

33Yet the prosodic cues in this example are a little more complex. In the first IP, with high pitch on the stressed syllable of “unfortunately” indicating the change in topic, this change is underlined by the fact that, after the high pitched stressed syllable, on the rest of the word and what follows “…unately over the next few days” the tone is maintained at a constant mid pitch level (circled in yellow). This prosodic feature involving two or three distinct steps, over the IP maintained at constant level is widely used35 and serves to structure the discourse. This phenomenon is referred to in Ashby and Maidment who point out that this use of what they call “key” “is sometimes very evident when newsreaders are changing from one news item to another”, (2005: 171). The same applies to the weather forecast.

3.2 Lexical and prosodic cues: commenting and projecting

  • 36 This IP could also be interpreted as a fall-rise with the fall beginning on “turning”.

34Beyond marking topic change, the choice of the word “unfortunately” implies that the following information will be negatively perceived, as suggested in the first part of this article. This enables the listener to predict to some extent what is to come, thereby facilitating comprehension. This type of comment on the information given is also present at the end of this extract, but this time with no lexical cue. The change in trend, “turning quite windy”, is said with a rising tone on “windy”,36 as if the forecaster was apologizing for the bad conditions he is predicting.

35In the last example, we find both lexical and prosodic cues working together to express a particular perspective. Again, the first word of the IP, “temperatures”, is associated with high pitch to show topic change; what follows is quite low but the words “up into double figures” express a positive trend: first as “up” indicates they are higher, and second, the idiomatic expression “double figures” is positively connoted, so a native speaker is guided in his/her interpretation of the precise information which is then given: “thirteen or fourteen” degrees Celsius.

Figure 8 - Example 13: Margaret Emerson 21/02/14 commenting and projecting

Figure 8 - Example 13: Margaret Emerson 21/02/14 commenting and projecting

36The next phrase “we could well see” implies a projection. The inclusion of “well” and the way that this word is highlighted with much higher pitch allows the forecaster to imply that the prediction is very probable, and in this instance, desirable. Indeed, it is immediately followed by figures which are to be understood as positive (“double figures”), even though in absolute terms they would not be perceived as such. The positive nature of the temperatures is clearly shown by the prosodic cues corresponding to the number “fourteen”, which is the nucleus of this IP. Indeed, the very high fall on “fourteen” reflects both perhaps the surprise of the speaker and certainly her enthusiasm.

3.3 Prosodic cues: pattern and variation

37The end of this extract, “by the time we get to Sunday” is said with much slower pace, thereby indicating the end of the topic, and in this case, the end of the weather forecast update. I shall now conclude by commenting on the table in Figure 9 which recapitulates the prosodic cues that have been found in this analysis. There is considerable variation in these cues, but some appear to be of particular significance and form a regular pattern.

Figure 9: recap of prosodic features

Figure 9: recap of prosodic features

38The first column includes a list of what is marked by prosodic cues in the weather forecast. The first two lines (in dark blue) concern the way forecasters structure the information they are transmitting overall, and how they go from one topic to a different one. The findings are relevant both in terms of topic change between IPs and within an IP. The most significant prosodic cue is high held pitch which is used to indicate a change as well as to highlight a particular topic within a given IP. Topic change is also frequently associated with pitch reset: low pitch at the end of the previous IP contrasts with high pitch at the beginning of the next IP. And within IPs, we often have slower pace at the end, as seen in the last example, or pitch declination over the IP, as seen in the first one.

39The next five items (in green) concern the informational contents themselves and what the forecaster is saying about them. High pitch is systematically used for highlighting, as expected, whether it be the information itself, who is concerned by it (the scope), or how important it is (degree). Likewise, high pitch is used to indicate trends, be they continuing or changing, as well as for markers of additional elements. Finally, the last line (pale blue) concerns the prosodic cues used by forecasters to comment on the information they give. This once again involves high pitch which is clearly the most significant prosodic marker in this type of discourse situation.

Conclusion

40In this analysis, I have tried to determine what the main characteristics of the weather forecast are, and there does indeed seem to be a regular pattern with a great many recurrent features, so they should be predictable. The forecast includes an enormous amount of, sometimes technical, information and we may wonder how significant it is that most people are usually just interested in one place or perhaps two, not in the whole forecast. Visual cues obviously help follow the information even if our interest is in one specific area, but happily, the specialist makes an effort to connect with the audience in other ways too.

41From a linguistic perspective, our analysis has shown that forecasters use a number of techniques which enable the audience to be receptive to important facts. First, as we have just seen, they use prosodic patterns to highlight information structure, as shown in the last table. The listener can therefore perceive when a new topic is being addressed, and focus on the most important facts as well as know who they concern, to what extent they will be affected, and how sure the forecast is. Widespread use is made of set expressions and collocations referring to the weather which are completely predictable for a native speaker and often have a reinforcing effect as the other word(s) is/are predictable in the context (“sunny spells and scattered showers”). Lastly, oral markers are often used, and an informal register including a number of idiomatic expressions so as to relate to the listener’s everyday way of speaking. The technical information is therefore presented within a more commonplace linguistic framework which is intended to make the listener feel more at home, thereby improving receptivity.

  • 37 Beyond obvious references like the different countries, certain regions like East Anglia, and espec (...)

42I should like to consider these findings in the light of the difficulties encountered by non-native speakers. To what extent could an understanding of lexical and prosodic characteristics help improve their comprehension of this notoriously difficult type of discourse? Their first difficulty is probably the fast pace. Greater familiarity with the prosodic cues which are used should enable them to understand better the overall information structure. This obviously needs to be associated with a knowledge of the specialist terminology, the lexical field in general and a good knowledge of the geography of the UK.37

43Yet, this is not enough for an understanding of the weather forecast… Paradoxically, the many idiomatic expressions, collocations, and informal words which are so prevalent in this type of programme and which make it easier for a native speaker, enabling them to predict to some extent what is to follow, are almost certainly a considerable source of difficulty for the non-native. The many words like “chilly”, expressions such as “on the cards”, collocations or set expressions like “scattered showers” and “outbreaks of rain” are unfamiliar and so add an extra difficulty. This must encourage us, as teachers, to work on oral English from a wide perspective, including prosodic, lexical and grammatical aspects, and highlight the many features which are specific to oral discourse.

Top of page

Bibliography

Primary corpus sources

http://www.metoffice.gov.uk (forecasts between 15/01/14 and 21/03/14)

https://www.youtube.com/watch?v=wp2L6pYH2w4 (Marco Petagna, Queens’s diamond weather jubilee forecast) 03/06/12

https://www.youtube.com/watch?v=OjqknxTDArc (Margaret Emerson, Queens’s diamond weather jubilee forecast) 04/06/12

http://www.youtube.com/watch?v=JvX-jOlIFds (Michael Fish 1979)

https://www.youtube.com/watch?v=IlR2QjQgzi4 (Phoebe Smith Blooper!)

https://www.youtube.com/watch?v=em4RCzrCpyU (Michael Fish interview, 29/10/13)

https://www.youtube.com/watch?v=54FKp4Ib4Kk (BBC Scotland Prince Charles reads the weather forecast, 10/05/12)

https://www.youtube.com/watch?v=56crttyLVJQ (Channel 4 news, 26/12/13, Marco Petagna interviewed)

Recordings by L2 LLCE students, University of Limoges, scripted forecasts, March 2014

References

Ashby, Michael and John Maidment. Introducing Phonetic Science. Cambridge: Cambridge UP, 2005.

Anthony, Laurence. AntConc [Computer Software] (Version 3.4.1). Tokyo, Japan: Waseda University, 2014. Retrieved 9 July 2014 from http://www.laurenceanthony.net/AntConc.

Boersma, Paul & David Weenink. Praat: doing phonetics by computer [Computer program]. Version 5.3.81, retrieved 3 July 2014 from http://www.praat.org/.

Crystal, David. Encyclopaedia of the English Language. Cambridge: Cambridge UP, 1995.

Crystal, David. Just a Phrase I’m Going Through: My Life in Language. Abingdon: Routledge, 2009.

Herment, Sophie. « Emphase prosodique et emphase syntaxique : le cas de « do » dans un corpus de parole naturelle. », CORELA - Numéros thématiques Parole. http://corela.edel.univ-poitiers.fr/index.php?id=839 (2010).

McCarthy, Michael and Felicity O’Dell. English Collocations in Use. Cambridge: Cambridge UP, 2005.

Moore Mauroux, Susan. “The Markers of Focalization in English”. In Cappeau Paul and Sylvie Hanote (Eds.). Focalisation(s). Rennes: PU Rennes, 2012 (59-82).

Sandré, Marion. Analyser les discours oraux, Paris : Armand Colin, 2013.

Silber-Varod, Vered and Loïc Kessous. “Prosodic Boundary Patterns in Hebrew: A Case Study of Continuous Intonation Units in Weather Forecast”. Proceedings of the Speech Prosody 2008 Conference, 2008 (265-268).

Wells, John. English Intonation, Cambridge: Cambridge UP, 2006.

Wichmann, Anne. Intonation in Text and Discourse: Beginnings, Middles and Ends. London: Longman, Pearson Education, 2000.

Background corpus

http://www.bbc.co.uk/weather/features/25658706 (Sixty years of BBC Weather presenting)

http://www.bbc.co.uk (BBC forecast for Europe: between 14/01/14 and 10/03/14

Channel 4 news and weather: between 17/02/14 and 20/02/14

https://www.youtube.com/watch?v=snqullFCu5Y (Met Office celebrates 150 years of forecasting for the nation)

https://www.youtube.com/watch?v=QuIV6BNsFd4 (Behind the scenes at the met office)

http://www.itv.com/news/london/update/2013-07-23/how-can-it-hail-in-the-middle-of-a-heatwave/ (Phoebe Smith interviewed by ITV)

Top of page

Annex

Appendix: selected extracts

Examples in bold are analyzed in the text

1/ We’re certainly not out of the woods yet / It does look as if / Very gradually in the next day or so / It’s going to turn cold again / over most parts of the country (MF 31/01/79)

2/ And for the remainder of the bank holiday weekend / We’ll see a mixed bag of weather across the UK / It’s a case of sunshine and showers on Monday / With further wet and breezy conditions / moving in from the south west as we head into Tuesday (MP 04/06/12)

3/ today it’s going to be a fairly disappointing day for many of us across the UK / lots of cloud around / wet for many of us as well through the day / but some of us may be a little bit lucky and see some brighter conditions later on (PS 15/01)

4/ False start: Outbreaks of rain as well / but some brighter spells pushing in to the west in particular later on / as well / aah I’ve I’ve / no I didn’t like that ‘cos I couldn’t remember if I’d actually said east or west … Try again (PS 15/01)

5/ then as we move through into tomorrow / we’re going to have an area of rain push up from the south / that’ll be bringing some really quite heavy rain / particularly to western areas through the course of the day / and generally further north / showers again / but some sunny breaks as well (PS 17/01)

6/ Temperatures are really going to be quite disappointing / Just a few degrees above freezing and feeling really quite chilly (PS 30/01)

7/ so pretty wet wash er going into the evening for much of the country but / a little bit dryer / the rain turning a bit more showery for a while over northern Ireland / and as you can see (ME 31/01)

8/ this afternoon thankfully is a respite for many of us / there’s some cloud and rain in the far south east / but for many it’s largely dry and bright / there’s just a scattering of showers around / some of them are heavy and a touch wintry over the highest ground (BY 10/02)

9/ still quite breezy and one or two showers but also a fair amount of fine weather too (ME 12/02)

10/ unfortunately over the next few days we’re like likely to see further outbreaks of rain across most parts of the UK / although on a more positive note it won’t be nearly as windy as it has been of late / and most places should just about escape a frost during the week ahead too (MP 17/02)

11/ unfortunately over the next few days we hold onto quite unsettled conditions across the UK / we’ll see further rain at times / it’ll turn quite windy / especially on Wednesday night into Thursday / although on a more positive note / it’ll stay fairly mild for the time of year / and not nearly as stormy as in recent days either (MP 18/02)

12/ temperatures are still pretty good for the time of year / once again and more especially towards the south of the UK / highs this afternoon of 10 or 11 Celsius / 11 in London / it’s 52 Fahrenheit / with fairly gentle winds out and about across the UK / when the sun does pop through / shouldn’t feel too bad out and about at all (MP 18/02)

13/ apart from a few a few showers there up in the north west / fair amount of sunshine / temperatures up into double figures / we could well see 13 or 14 by the time we get to Sunday (ME 21/02)

14/ A couple more days of fine settled weather across southern parts of the UK are on the cards / but further north we already have some unsettled weather / and that’ll be spreading to the whole of the UK through the week / turning wet and windy for many of us and feeling notably cooler by the weekend (PS 18/03)

Top of page

Notes

1 http://www.metoffice.gov.uk forecasts between 15/01/14 and 21/03/14 involving three male speakers and two female speakers, all native speakers. Two other Met Office recordings have been included involving the same speakers, from the Queen’s diamond Jubilee in June 2012: https://www.youtube.com/watch?v=wp2L6pYH2w4 and https://www.youtube.com/watch?v=OjqknxTDArc. Details concerning the number of forecasts analysed, duration and pace per speaker and are given in Figure 2.

2 http://www.youtube.com/watch?v=JvX-jOlIFds Michael Fish 1979.

3 Certain TV forecasters, as for example Liam Dutton (Channel 4 news and weather, February 2014) present the weather forecast in a more engaged way, but the overall characteristics found in this analysis are still relevant. The visual element present, however, involves a number of specificities which distinguish it from a radio forecast.

4 BBC Scotland Prince Charles https://www.youtube.com/watch?v=54FKp4Ib4Kk, and twenty L2 LLCE English students from the University of Limoges who each prepared a script for a weather forecast (from icons on the BBC website) which they then read. Five students are native speakers.

5 See interviews with Marco Petagna https://www.youtube.com/watch?v=56crttyLVJQ and Michael Fish https://www.youtube.com/watch?v=em4RCzrCpyU.

6 Our analysis of the lexicon was based on findings using the concordancer, AntConc (Anthony 2014).

7 This phrase was used by Hélène Chuquet, University of Poitiers, as an object line in her email enquiring about this paper, capturing the essence of its objectives, including both the idea of predictability and variation. My thanks to her also for re-reading this article, and for her constructive comments.

8 Crystal speaks of surreptitiously recorded conversations, where the pace reaches “well over 300 syllables a minute at times instead of the sedate 200 or so used on the radio.” (Crystal 2009: 103). Bearing in mind that my figures refer to words and not syllables per minute, the weather forecasts under study are all considerably faster than what Crystal expects in a radio programme.

9 One of the objectives of the course is reading aloud from a script.

10 Channel 4 news : https://www.youtube.com/watch?v=56crttyLVJQ

11 These performance errors were made by Phoebe Smith in the 19/03/14 and 6/02/14 updates. This contradicts Silber-Varod and Kessous (2008) who consider that errors of production are not tolerated, but this does not invalidate their main point which is that there are no truncated boundaries.

12 Compare examples 3 (the final version) and 4 (false start) in the list of examples in the Appendix. For further examples showing the forecast is not scripted (and giving insight into what is going through the forecaster’s mind during the presentation), consult https://www.youtube.com/watch?v=IlR2QjQgzi4 (Phoebe Smith Blooper! weather and climate change) where she reminds herself not to forget Scotland!

13 The corpus has 1118 word types, for a total of 10 303 word tokens in all. The statistical analysis was made using AntConc (Anthony, 2014)

14 “The detail” refers systematically to the satellite picture, never to the more common sense.

15 The pairs “south-southern”, “west-western” “north-northern” (without occurrences of “northern Ireland”) occur between 100 and 120 times, so are among the most frequently occurring words in the corpus (but there are fewer than 60 tokens for “east-eastern”).

16 See Example 1, Michael Fish. The phrase is also used by Aidan McGivern (26/01/14).

17 Markers of degree include the following (number of tokens): “a bit” (20), “a little bit” (23), “pretty” (18), “fairly” (27), “quite” (54). Other examples of informal register are: “the odd patch” (3), “a fair amount” (3).

18 In addition to being colloquial, both the adjective « chilly » and the adverbs modifying it seem imprecise and unscientific, expressing a somewhat subjective, vague judgment.

19 Colloquial expressions are often used: “a mixed bag” (Example 2), “when the sun does pop through” (Example 12), “on the cards” (Example 14).

20 My thanks to Ramon Marti-Solano, University of Limoges, for his helpful suggestions on collocations and set phrases, and on concordancer techniques in general.

21 If we hear “outbreaks”, this automatically implies “rain”.

22 It is interesting to note that in 1979, Michael Fish uses “you” rather than “we”. His use of modality is also different, reflecting the greater degree of uncertainty regarding the prediction, but this will not be studied in this article.

23 Other examples of this combination of using an oral register and giving a point of view are: “the best of the sunshine”, “not too bad”. The prediction itself is often worded using an oral expression: “we could well see”, which occurs 8 times. See Figure 8 for an analysis of prosodic features.

24 Pitch corresponds to the acoustic feature, fundamental frequency (F0), measured in Hz, as seen in the figures created thanks to PRAAT. These also show duration and intensity although these features have not been specifically analyzed in this paper.

25 Variation between speakers also needs to be taken into account as different speakers make use of different prosodic markers. Moreover, there are stylistic differences between speakers, including the way in which they present themselves or end the update.

26 When making a transcript of the weather forecast, I realized how circular the information content is, although the graphic representations allow us to follow the chronology. Likewise, the (same) information tends to be reformulated: presumably, the forecaster tries to avoid saying exactly the same thing for stylistic reasons (and one may wonder whether this has any connection to the number of forecasts they give per day). Crystal (1995: 385) refers to “formulaic phrasing” which enables them to produce fluent, spontaneous speech respecting the time constraints.

27 He lays considerable emphasis on the modal “could” in “where it could lead to difficult conditions on the road”, speaking of the potential effects of heavy rain. As part of the weather forecast, it is unlikely that this modal would be stressed, although it would retain its full form.

28 The pitch contour (F0) is circled in yellow.

29 The same prosodic markers apply to the next topic which is introduced by “if we take a look….”, which points us in another direction, and the change in topic is again reflected in a (longer) pause followed by a high pitch reset.

30 See Herment (2010) and Moore Mauroux (2012) for a more detailed analysis of prosodic prominence within contexts of emphasis/ special focus.

31 Prosodic prominence also occurs in indications of scope (“most”, “many”, “some”) or degree (“pretty”, “fairly”), allowing the forecaster to convey the relative significance of the information given.

32 We frequently find that after marking and highlighting the idea of continuation, the information we already know is not prominent (cf. Wells 2006 “given information”).

33 The nucleus of the IP is circled in black (both the pitch contour and word) here and in Figure 7.

34 In this case, “too” is the nucleus of the third IP and so the new additional information is surrounded by markers that echo each other: “also” and “too”.

35 This is a stylistic feature which is frequently used by some forecasters, including Marco Petagna.

36 This IP could also be interpreted as a fall-rise with the fall beginning on “turning”.

37 Beyond obvious references like the different countries, certain regions like East Anglia, and especially the names of mountain ranges like the Pennines, are a definite comprehension obstacle.

Top of page

List of illustrations

Title Figure 1: graphic representations
URL http://anglophonia.revues.org/docannexe/image/755/img-1.png
File image/png, 20k
Title Figure 2: the corpus, pace for professional weather forecasters
URL http://anglophonia.revues.org/docannexe/image/755/img-2.png
File image/png, 23k
Title Figure 5 - Example 3: Phoebe Smith 15/01 – discourse markers and pitch
URL http://anglophonia.revues.org/docannexe/image/755/img-3.png
File image/png, 96k
Title Figure 6 - Example 9: Margaret Emerson 12/02/14 - the focusing function
URL http://anglophonia.revues.org/docannexe/image/755/img-4.png
File image/png, 83k
Title Figure 7 - Example 11: Marco Petagna 18/02/14 – focusing and commenting
URL http://anglophonia.revues.org/docannexe/image/755/img-5.png
File image/png, 101k
Title Figure 8 - Example 13: Margaret Emerson 21/02/14 commenting and projecting
URL http://anglophonia.revues.org/docannexe/image/755/img-6.png
File image/png, 192k
Title Figure 9: recap of prosodic features
URL http://anglophonia.revues.org/docannexe/image/755/img-7.png
File image/png, 18k
Top of page

References

Electronic reference

Susan Moore Mauroux, « Patterns and Variation in the Weather Forecast: Can Prosodic Features be Predicted Too? », Anglophonia [Online], 21 | 2016, Online since 01 July 2016, connection on 19 August 2017. URL : http://anglophonia.revues.org/755 ; DOI : 10.4000/anglophonia.755

Top of page

About the author

Susan Moore Mauroux

Université de Limoges, CERES EA 3648
susan.moore@unilim.fr

Top of page

Copyright

Licence Creative Commons
Anglophonia – French Journal of English Linguistics est mis à disposition selon les termes de la licence Creative Commons Attribution - Pas d'Utilisation Commerciale - Pas de Modification 4.0 International.

Top of page
  • Logo Presses universitaires du Midi
  • Revues.org