Normative Generics and Norm Breaching – A Questionnaire-Based Study of Parent-Child Interactions in English

Abstract The present paper focuses on the phenomenon of normativity and genericity in language and cognition. More specifically, it investigates the use of normative generics, which are generalizations that state an ideal norm for a given category, in the context of norm breaching in parent-child interactions in English. This issue is researched by means of a specially designed questionnaire including 8 norm breaching parent-child interactions, which has been completed online by ca. 70 English-speaking female respondents. The paper uses qualitative and quantitative methods to address two specific research issues. First, it compares the frequency of use of normative generics in norm breaching situations vis-à-vis the use of other types of normative linguistic expressions. Second, it analyses selected factors that are believed to favour the use of normative generics, including interactive openness of a given situation, norm salience, and perceived norm importance. Moreover, the paper sketches an explanatory model of normative generics that draws upon insights from the Conceptual Metonymy Theory, Construction Grammar, and Dual System Theory.


Introduction
Genericity and normativity are phenomena that have long been discussed by scholars of various persuasions including philosophers, legal theorists, psychologists and linguists (Bicchieri, 2006;Brożek, 2012;Krifka et al., 1995;Paulus & Schmidt, 2018). Genericity concerns the relationship between a property and a kind (Nickel, 2016), whereas normativity designates what agents ought to or ought not to do (Star, 2018). The present study lies at the intersection of these two broadly defined research areas by focusing on normative generics, that is, statements such as boys don't cry, which implicitly indicate a norm that applies to a given group. Unlike descriptive generics, normative generics are often ignored in the generics literature, where they are thought to be aberrant cases or 'not real generics' (McConnell-Ginet, 2012).
Descriptive generics are generalizations such as tigers are striped, mosquitoes carry malaria or ducks lay eggs, which express claims about a kind rather than an individual, although they are not universal or exceptionless. There are two main theoretical approaches to descriptive generics: linguists (Krifka et al., 1995;Langacker, 1997Langacker, , 1999 treat such generics as quantificational, whereas psychologists treat them as non-quantificational (Gelman, 2010;Leslie, 2007Leslie, , 2008Leslie, , 2012. Normative generics, on the other hand, have received little scholarly attention so far. Two models proposed by Cohen (2001) and Leslie (2015) deserve special mention because of their detailed treatment of normative generics. These two accounts, however, offer radically different approaches to the phenomenon in question (see section 4 for a brief discussion of the two models).
Following Brennan, Eriksson, Goodin, and Southwood (2013), norms are understood as accepted rules or normative principles. The use of the term "norm" is not, however, without problems as it has acquired different meanings for different researchers. 1 For instance, according to Wedgwood (2018, p. 2), there are three senses of the word norm that are in use among philosophers: (1) a 'model' or 'standard' that helps to navigate people's thinking or behaviour, (2) 'rules of obligation' discussed in the philosophy of law, and (3) a 'general principle' according to which rational agents should act.
Drawing upon insights from philosophy (Brennan et al., 2013;Leslie, 2015), psychology (Schmidt & Rakoczy, 2018) and linguistics (Langacker, 1999;Radden, 2009), the aims of this paper are twofold: (1) to investigate types of linguistic expressions that are used to communicate that a given situational standard has not been met in selected parent-child interactions and to identify the frequency of normative generics among them and (2) to discuss selected situation and norm-related factors that favour the use of normative generics. These two questions will be answered by analysing the results of a survey conducted among female native speakers of American English. Finally, the conclusion contains a concise outline of an explanatory model of normative generics that is currently being developed by the authors. norms can be divided into two types: practical and theoretical (or epistemic). Simply put, the former regulate human action and the actions we should take, while the latter regulate human belief and the reasons for what we should believe in (Glüer & Wikforss, 2018).
Another classification claims that moral norms "are clusters of essentially practice-independent judgments," while social norms are "clusters of practice-dependent normative judgments" (Brennan et al., 2013, p. 72). As an illustration of what makes moral and social norms markedly different, the authors give the example of a social norm among Oxford dons that requires dons to pass the port to the left. If the don is asked to justify why one must pass the port to the left, they might respond by saying "That's just the way it's done around here". This kind of answer seems natural and compatible with the question. However, appealing to 'what is done' when answering a question about a moral judgment such as "Why do you think that one must not commit murder?" seems unnatural and incompatible with the question. Thus, it appears that 'what is done' can be a suitable reply to a question involving a social norm but not a moral norm.
Having these categories at hand, let us now turn to the explanatory potential of norms. Norms are usually recalled to explain why things are done the way they are done. For instance, when a chess player makes a wrong move, a student cheats in the exam or a sister kicks her younger brother, norms might be called up to explain why a given move is not allowed, kicking is not a good behaviour or cheating is unfair to other students. There are several ways in which norms can explain our behaviour. Following Brennan et al. (2013), we focus on three such ways: (1) norm following, (2) norm conforming, and (3) norm breaching. Two features of social norms make them particularly suitable for norm following. First, only a significant number of members of a given group decide whether a rule is a social norm (e.g. a significant number of Oxford dons decides that the port must be passed to the left). Second, social norms "are constituted by practicedependent normative attitudes and create a kind of social accountability" (Brennan et al., 2013, p. 260). This is to say that social norms, which are grounded in social practice 3 , require us to follow a norm or otherwise suffer the consequence of failing to do so. The lack of table manners (i.e. failing to comply with a norm), might entail not getting invitations to dinner parties. On the whole, one can be said to follow a norm if the norm has been internalised. This means a norm is treated as a non-instrumental reason for an action, i.e. a norm is followed because it is a norm. However, mastering table manners purely for the sake of being invited to dinner parties is not an example of norm following but norm conforming. Unlike norm following, norm conforming involves treating a norm as an instrumental reason for action (i.e. to externalise norms).
Finally, let us focus on norm breaching, which is the focus of the present study and is defined as acting contrary to what is required by a norm (Brennan et al., 2013). It is generally agreed that people deliberately breach norms; however, it would seem that they hardly ever ignore a norm altogether. Instead, they might, e.g. pause and think about the consequences of norm breaching or the reasons for having the norm in the first place. Some types of norm breaching include avoiding a norm, acting opposite to a norm or pretending to comply with a norm. Overall, we breach norms because complying with a norm (following it or conforming to it) is too costly.

Normative linguistic expressions
We have a repertoire of linguistic means at our disposal that can be used to express that a particular norm/rule is to be followed or that a norm/rule has been breached. These normative linguistic expressions include, e.g. deontic verbs and adjectives, imperatives, discourse particles and normative generics 4 . Deontic modalities -usually defined in terms of permission and obligation -comprise such auxiliary verbs as must, have to, may, can, should and ought to as well as the adjectives obligatory, permissible and impermissible (Huddleston & Pullum, 2002; but see e.g. Nuyts, Byloo, & Diepeveen, 2010 for some redefinition of the category). On the other hand, imperatives are statements such as go to bed or please close the window, which are used to express a command, an order or a request to be followed by one's interlocutors (Huddleston & Pullum, 2002). The category of discourse particles includes such examples as well, which, according to Bolinger (1989, p. 321), can imply a notion of conformity with a norm. Bolinger (1989, p. 331) illustrates this use of well by citing the following example: "Ronald is chasing skirts again. Well! And him with a wife and six kids!". The last category comprises normative generics such as boys don't cry, which seem to be yet another device fit for communicating that norms and rules have to be followed or that they have been breached. 5

Generics, normative generics and normative force
Generics are generalizations that are ubiquitous in everyday life. We rely on them to convey information to one another. Generics can assume a number of syntactic forms in English: a lion has a bushy tail (indefinite singular), lions have bushy tails (bare plural) and the lion has a bushy tail (definite singular). They can be characterised by their reference to several dimensions: (1) temporal unboundedness, (2) law-like nomic character, (3) association with dispositions, (4) resistance to contextual restriction, and (5) exception tolerance (for a discussion of these, see Lazaridou-Chatzigoga, 2019). There are two main approaches to generics: the quantificational view and the non-quantificational view. The former comprises at least several theoretical models: a modal approach (Krifka et al., 1995), a normalcy approach (Pelletier & Asher, 1997;Nickel, 2008) or a probabilistic approach (Cohen, 1999), while the latter includes the generics-as-default view. This approach assumes that generics are innate and involve a default mode of thinking (Leslie, 2007(Leslie, , 2008(Leslie, , 2012. Normative generics, that is, statements such as boys don't cry, a girl plays with dolls, or a woman puts family before career have not received systematic attention in the literature thus far. The communicative import of these generics is that 'boys should refrain from demonstrating their emotions and vulnerability', 'girls should not play with cars or plastic guns' and 'women should care more about their families than careers'. Thus, on the one hand, normative generics explicitly communicate one thing ('boys do not cry') but on the other, they implicitly communicate another ('boys should refrain from crying'). We can distinguish several dimensions of normative generics: (1) they have a normative or hortatory force, (2) they endorse norms and (3) they are assigned different truth conditions than descriptive generics (for a discussion of these, see Hesni, 2019). There are two competing theoretical models of normative generics proposed by Cohen (2001) and Leslie (2015). 6 The former model proposes that normative generics require different logical forms than descriptive generics such as ducks lay eggs. Under his approach, normative generics are explained in terms of a rule 'if x is a boy/girl, x doesn't cry/plays with dolls' and an unpronounced predicate 'is in effect'. The latter model assumes that there is no need to postulate a descriptive vs. normative account of generics on the ground that descriptive/normative duality lies within the nominal itself via lexical polysemy in the subject position. More specifically, her account is centred around the notion of dual character concepts (Knobe, Prasada, & Newman, 2013) and of normative force.
Normative generics (as well as norms) seem to possess a normative or hortatory force, which is understood as a "binding force" (Schmidt & Rakoczy, 2018) or an "obligation" (Leslie, 2015), that propels us to do what is required of us. More specifically, normative generics implicitly urge us to act and behave in a certain way or, to put it another way, they urge us to endorse a certain norm. Unlike a physical force such as gravity, normative force binds us to follow a norm, but we may decide to breach it.

Methods
In order to investigate the role of normative generics in parent-child interactions, the present study made use of a specially designed questionnaire (see Appendix for details). The questionnaire was divided into two parts. The introductory (warm-up) part consisted of two short descriptions of everyday situations and asked the respondents about their familiarity with such situations and their potential reactions to them. This part was designed to expose the respondents to conceptually similar stimuli in order to facilitate their thinking about norm breaching in parent-adult interactions and was not the subject of the analysis in the present study. The second part of the questionnaire comprised 8 short descriptions 7 of selected everyday situations that involved parent-child interactions and potential norm breaching 8 . The respondents were asked to briefly describe what their reactions to potentially norm breaching behaviour of a child would be in a given context. The contexts included, among others, a little girl kicking her brother, and a little boy asking his mother to paint his nails. The study was carried out by means of the online survey platform Survey Monkey and also included a set of socio-demographic questions. A total of 71 to 74 (depending on the question) female individuals 9 participated in the study, which lasted about 10 minutes for each. All the participants reported being native speakers of American English.

Results
The analysis of open-ended responses provided by the participants in the study demonstrated that the use of normative generic expressions significantly depends on the type of situational context provided in the question.
As far as the first situation in the questionnaire is concerned (You are doing some important paperwork. Your little son, who is playing next to you, gets louder and louder with every moment. Please write briefly in one sentence what you would tell him in such a situation), in which the norm breaching consists in playing too loud and therefore interrupting a parent, none of the 71 respondents employed a normative generic expression. Amongst the most frequent answers were kind requests that a child act quietly or leave the room combined with explanations or promises to do something together afterward (54.92% of all the responses), e.g. "Please play quietly", "Would you please play quietly", and explanations followed or preceded by promises or other offers (15.49% of all the responses), e.g."Give me a min and I will play with you".
A similar lack of normative generics is also observed in situation no. 8 (You enter the living room and see that your little daughter has left her toys scattered all around the room. Please write briefly, in one sentence, what you would tell her in such a situation.), in which the norm breaching consists in a child not respecting cleanliness standards at home, in particular, not tidying up the living space after playing there. In this situation, the 74 surveyed respondents opted for a wide range of reactions, the most frequent of which being kind requests to clean up (28.37% of all the responses), e.g. "Please clean up your toys", and requests or proposals to clean up together (21.62% of all the responses), e.g. "Let's put your toys away together". Three other parent-child interactions presented in the questionnaire that showed a conspicuously low number of normative generic responses concerned situations no. 2, no. 4 and no. 6. In situation no. 2 (Your little son eats dinner with you at the table and is sitting crooked. This starts to irritate you and you want to reprimand him. Please write briefly, in one sentence, what you would tell him in such a situation.), the norm breaching is related to table manners, in particular, a preferred way of sitting at the dinner table. Here, the most frequent responses given by the 72 participants included either signalling that this norm is not important for them (31.94% of all the responses), e.g. "Don't care if he's sitting crooked", or kind requests that were sometimes followed by the explanations (29.16% of all the responses), e.g. "Please sit up straight". As far as normative generics are concerned, there was only one response ("Honey, at the table we can't sit like that") in which the pronoun we could be interpreted as both referring to a particular family or people in general. Despite the fact that there was a significantly low frequency (1.38% of all the responses) of normative generic expressions in this context, it is worth mentioning that the respondents quite frequently (18.04% of all the responses) used other grammatical constructions (kind requests and imperative sentences) that included evaluative vocabulary (discourse particles) indicating normativity (see Bolinger, 1989). These included adjectives such as properly, appropriately, right, and correctly.
As for situation no. 4 (Your little daughter sits with you at the dinner table. You can see that she is not interested in what she has on her plate and she starts to play with the food. Please write briefly, in one sentence, what you would tell her in such a situation.), in which the norm breaching is related to both table manners and food taboo rules that require a child to behave in a particular way towards food while eating with others, the normative generic was only used once in 74 responses as part of a longer utterance including also another type of reaction ("I might ask if she is feeling well, and if she is OK, I'd tell her that we don't play with our food."). In this case, as in situation no. 2, the normative generic expression "We don't play with our food" included the pronoun we to refer to either that particular family or people in general. Similar to situation no. 2, the responses also included explicit use of a normative and evaluative adjective ("It is not appropriate to play with your food"). However, this only occurred once among all the responses. On the other hand, the most frequent responses to this situation were conditional imperatives, which appeared in 36.48% of the responses, e.g. "If you're finished, you may leave the table", and questions about appetite or food quality, which appeared in 33.78% of the responses, e.g. "Do you not like your dinner?".
In situation no. 6 (Your little son fell down while playing on the playground and slightly scratched his knee. He runs up to you all in tears. Please write briefly, in one sentence, what you would tell him in such a situation.), in which the norm breaching behaviour consists in not meeting the idealised gender standard of how men and boys should react to pain or failure, the normative generic reaction also appears only once in the 74 responses. It is used as part of a longer utterance: "Hey, hey, let's take a look, not much blood, we better go check the sidewalk and see if you chipped it. LOL Big boys don't cry". In this case, however, the normative generic is hedged by the expression LOL, which stands for 'laughing out loud' and is commonly used in texting and online communication. It is believed that in this particular context, the combination of LOL with big boys don't cry is meant to signal the ironic use of the normative generic expression. The most frequent responses in this situation include comforting declarative sentences (used in 55.4% of the responses), e.g. "You're OK, it's only a scratch", "Let me kiss it and it will be better soon"; let me see/take care-type declarative sentences (appeared in 28.37% of the responses), e.g. "Let me look at your knee"; and questions about the son's condition or needs (found in 21.62% of the responses), e.g. "Are you ok?", "What can I do to help you feel better?". These questions are often combined with one another and with other reactions to form longer utterances.
The use of normative generic expressions has been significantly more frequent in the remaining three situations. In situation no. 3 (Your little daughter kicked her younger brother in your presence. Please write briefly, in one sentence, what you would tell her in such a situation.), in which the norm being breached relates to the use of physical force among siblings, the 72 respondents often provided longer responses that combined at least two types of reactions, e.g. an imperative sentence (the most frequent expression that appeared in 50% of the responses) plus requests to apologise. Normative generic expressions with the pronoun we such as "We don't kick other people" or "We do not use our bodies to hurt others" were found in 20.8% of the responses. Besides normative generics, the respondents frequently used expressions explicitly indicating the undesirable behaviour and including normative and evaluative vocabulary (see also situation no. 2 above and Bolinger, 1989), e.g. "Kicking is not a nice thing to do" or "Kicking someone is never acceptable behaviour". These types of reactions to norm breaching were found in 33.33% of all the responses.
A high number of normative generic responses were also found in situation no. 7 (You are painting your nails while sitting at the table. Your little son comes up to you and asks if you could paint his nails too. Please write briefly, in one sentence, what you would tell him in such a situation.), in which the norm breaching consists in not meeting the idealised standard of male physical appearance. In this context, normative generic expressions such as "Boys don't wear nail polish", "Boys only paint their nails in clear polish" or "Boys don't paint their nails" were used to signal the prohibition of a given behaviour by 10 (13.5%) of the 74 respondents. It is noteworthy that modified (or non-prototypical) normative generics such as "Boys don't usually paint their nails" were also used to precede the permission of a requested behaviour, which was the case in 4 (5.4%) of all the responses. Although normative generics have made up in total 18.9% of all the responses, they were not the most frequent reactions to this situation. First place was taken by permission expressed by means of various grammatical constructions, which were used by 53 participants (71.62% of all the responses).
The highest number/percentage of normative generic responses in the questionnaire was observed in situation no. 5 (You make coffee to drink it at breakfast with your family. Your little daughter asks if she could also drink coffee. Please write briefly, in one sentence, what you would tell her in such a situation.), where the norm being breached is related to a food taboo that classifies beverages into adult and non-adult categories, the former being purportedly dangerous for children. In this context, normative generic ex-pressions such as "Coffee is for adults", "Coffee is only for grownups", "Coffee is a grown-up drink", or "Kids don't drink coffee" were used by 23 participants (31.08% of all the responses), who employed them to signal their prohibition of this type of behaviour. On the other hand, the most frequent response was permission, which was expressed in numerous stylistic and grammatical ways and comprised 54.5% of all the responses.

Results summary and discussion
The present study set out to investigate the use of normative generic expressions in parent-child interactions that involve different types of norm breaching. The analysis of the questionnaire data shows that in only 3 out of 8 situations did the respondents use normative generics to signal norm breaching with a conspicuously high frequency (18.9% in situation no. 7; 20.8% in situation no. 3; 31.08% in situation no. 5). As far as the other investigated situations are concerned, normative generics were either not used at all (see situations no. 1 and no. 8) or used only once or twice (see situations no. 2, 4, 6).
It is believed that in order to elucidate these results and to better understand the specificity of normative generic expressions, several interplaying factors need to be addressed. 10 First, it is argued that the use of normative generics in a given parent-child interaction can be influenced by the interactive openness of that situation. In other words, it can depend on an array of available and conceivable reactions to a child's norm breaching behaviour (see Fitch & Sanders, 2005;Sidnell, 2010 for extensive discussions on the multiple facets of communicative situations). For instance, situations no. 5 and no. 7, with a high frequency of normative generics, can be characterised as having a limited number of possibly conceivable reactions on the part of a parent. In these cases, the respondents can either prohibit or allow something straightforwardly or in a more elaborate way, e.g. by using a normative generic expression. On the other hand, situations no. 1 and no. 8 may invite a parent to use a wider range of conceivable reactions to norm breaching such as promising, offering, proposing, or asking questions. The potential role of this factor is supported by comparing the number of types of responses provided by the participants in the questionnaire analysed. In situations no. 1 and 8, where normative generics are not used at all, the participants use as many as 7 response types. A similar tendency can be observed in situations no. 2, 4, and 6, where the number of response types, which are also combined to form longer ut-terances, also reaches 7. On the contrary, in situations no. 7 and no. 5, the number of potential reactions is lower. In the first one, we can find only 4 response types, and in the second, there are only 3 response types. 11 Thus, it is argued that the interactive openness of a given situation can impact the use of normative generics: the likelihood of a normative generic being used is lower when the number of conceivable reactions to norm breaching is higher. 12 Another factor that should be taken into consideration is norm salience (see Ghazizadeh, Griggs, & Hikosaka, 2016;Giora, 2003;Schmid & Günther, 2016, for various perspectives on salience used in cognitive linguistics and psychology). For the purposes of this study, norm salience is understood as the easiness with which a participant can recognise that a given situation is an instance of norm breaching and the easiness with which he/she recognises the content of the norm being breached. The situations without, or nearly without, normative generics are similar in that they can be seen as instantiating norms that are not particularly salient for the participants of this study, e.g. play quietly while your parent is working, do not play with food while sitting at the table. This could be either due to the situations themselves, for instance, their structural features, the shortness of the descriptions, or to the participants themselves, who might not be able to express the content of a norm in an explicit way. 13 On the other hand, in situations in which the use of normative generics was more frequent, e.g. kicking a brother, the norms being breached can be seen as more salient for the participants. This not only means that the participants are capable of quickly recognising what kind of norm is at play upon reading a short description, but they are also capable of expressing the content of a norm in a shorthand manner by means of dedicated linguistic normative expressions, i.e. normative generics.
Thirdly, the use of normative generics can be affected by the perceived importance of the norm. 14 This factor refers to the status the participant ascribes to a norm in a given situation. It depends on both subjective axiological preferences, e.g. some people may consider some norms to be more important than others; and axiological conventions present in a given community, e.g. some subjects may treat a given norm as important because of its social significance in or for a group they are members of, not because of their own convictions (see Brennan et al., 2013, for a general discussion of this subject). Moreover, the ascription of a given status to a norm is directly connected with the reactions that can result from it. One of these possible reactions is exemplified in extreme in situation no. 2, where the potential norm breaching, i.e. sitting crooked at the table, was not considered as an instance of norm breaching by 31.94% of the respondents. In other words, a large proportion of participants denied the existence of such a norm and therefore did not react to it having been breached. Nonetheless, this linguistic reaction, which could be seen as ascribing a zero status (non-existence) to a norm, only indicates one of many attitudes to the importance of the norm. These include, for instance, recognising the norm's status but not reacting to it having been intentionally or unintentionally breached or acknowledging the norm's importance and ergo responding linguistically. The results show that the surveyed participants express the latter by means of a wide range of linguistic expressions including kind requests, conditional sentences, imperatives, and obligations. As far as normative generics are concerned, it is argued that the likelihood of using a normative generic expression can be influenced by the perceived importance of a norm at hand. In other words, the situations in which normative generics were quite frequently used could be conceived as referring to more important norms, i.e. those that are axiologically more urgent to address, which, in turn, in a way invite to be explicitly communicated to a child. This factor could contribute to explaining the high frequency of normative generics in situations no. 3, 5, and 7, in which norms being addressed related to physical violence, dangerous food, and gendered norms of physical appearance (see Geary, 2010;Meyer-Rochow, 2009;Shackelford & Weekes-Shackelford, 2012, for discussions of the role of these factors in human evolution and culture; see also Leslie, 2015, for a general discussion of normative generics and gender), vis-à-vis those with supposedly lower status such as playing too loud or sitting crooked at the dinner table (see Bourdieu, 1984;Fischer, Karl, & Fischer, 2019;Pilcher, 2012, for discussions of some factors that could explain that).
In conclusion, as seen above, elucidating the use of normative generics in parent-child interactions requires taking into account several interconnected factors that are related to the characteristics of given situations and norms. The present discussion touches upon only three of them and does not claim to exhaust other potential factors that are at play, e.g. the sociodemographic characteristics of the participants. On the other hand, other limitations of the present study can be related to its general design. Being an exploratory, questionnaire-based study, it is open to a range of potential, usual-suspect problems including dishonest answers, participants' difficulties interpreting the questions and situations presented, as well as the participants not being able to fully express their emotional responses or feelings (see Andres, 2012, for a general discussion of the specificity and limitations of questionnaire-based studies).
Despite the above-mentioned limitations, it is argued that, based on a significant sample of responses and ecologically valid examples, we can draw some tentative conclusions and construct some hypotheses that could be of importance for future research and for developing a model of normative generics. Firstly, the presence of normative generics appears to be conditioned by at least three interplaying factors: (1) low interactive openness of a potentially normative situation, e.g. question-answer interactions are preferred; (2) high norm salience, e.g. a norm that is conspicuous, clearcut, and easy to formulate is preferred; (3) high social and individual norm importance, e.g. norms that are highly significant for the participants and their groups, such as those related to basic biological and social welfare issues, are preferred. Secondly, it is argued that the presence of these factors can promote a fast normative interpretation of a given situation, which, in turn, favours the use of a normative generic expression. Thirdly, it is claimed that the use of normative generic expressions in parent-child interactions, but possibly also in other interactive contexts, can be accounted for by combining insights from Dual System Theory (Kahnemann, 2011), Construction Grammar (Hoffmann & Trousdale, 2013), and Metonymy Theory (Littlemore, 2015). In this account, the use of normative generics in order to react to a particular norm breaching event would be seen as activating a fast and automatic response path (System 1 -intuition), which is couched in a dedicated, ready-made grammatical construction not allowing for many modifications, and which is motivated by a metonymic shortcut (stand-for or through-connection) that allows interlocutors to communicate with fewer words what would otherwise require expressing a range of interconnected propositions related to a given norm.

Funding
The project is financed from the grant received from the Polish Ministry of Science and Higher Education under the Regional Initiative of Excellence programme for the years 2019-2022, project number 009/RID/2018/19, the amount of funding 8 791 222,00 zloty. pro-attitudes (or beliefs about the presence of pro-attitudes) towards the relevant response that are a matter of (common) knowledge among themembers of the group".
4 An in-depth discussion of normative expressions is beyond the scope of the paper. 5 Note that normative generics lack any explicit evaluative terms such as should or ought (Hesni, 2019).
6 But see also Hesni (2019) for a metalinguistic theory of normative indefinite singulars. 7 The number of situations was limited to 8 in order to avoid survey fatigue and to increase the survey completion rate. 8 The parent-child interactions used in the questionnaire are based on some every day, ecologically valid, personal experiences of the authors, their families, and their colleagues. They were not selected according to any specific criteria.
9 Due to the potential influence of gender on norm breaching reactions, the survey was limited to female respondents only. Male respondents will be researched in other questionnaire-based studies. 10 Due to space limitations, this section discusses only some, selected situation and normrelated factors that are believed to influence the use of normative generics. 11 This downward trend is not confirmed by situation no. 3, which, despite having a relatively high number of normative generics (20.8%), involves 6 different reaction types that are combined in various ways. It is believed that this exception can be explained by other factors, such as the importance of the norm represented. 12 Verifying this and other hypotheses related to the presented factors requires further empirical testing. 13 It is important to mention that the norm does not need to be salient for a parent to react to norm breaching. It is logically possible for a parent to linguistically prohibit or disapprove of some behaviour even though he/she is not capable of explicitly and precisely expressing what he or she is sanctioning (Brennan et al., 2013). 14 It is believed that the notion of norm salience is influenced by its perceived importance. It could be hypothesized that the higher the perceived importance of a norm, the more salient the norm. This could, in turn, increase the probability of using normative generics in order to react to its breaching.