The Relational Approach and the Family

The article is an illustration, from a sociological viewpoint, of the epistemological and methodological scope of the relational approach to the study of the family. From this perspective (Section 1), the family can be identified as a social relationship of full reciprocity between sexes1 and generations, to be neither replaced nor confused with anything else (family genome), thus highlighting the advantages of the relational approach compared to other contemporary approaches, defined as relationist/relationistic. Section 2 focuses on the Added Value of the Relational Method from logical, methodological and empirical levels. The main traits of the family research are to consider the family as a unicum and its relational and historical nature. The AGIL scheme helps to analyse the family as a sui generis relationship by identifying its constitutive resources (A), aims (G), norms (I) and latent cultural model (L).


Introduction
The present contribution is meant to illustrate, from a sociological viewpoint, the epistemological and methodological scope of the relational approach to the study of the family.The paper consists of two sections: a more substantial one (1), aimed at highlighting the advantages of the relational approach as to other contemporary approaches, defined as relationist; a shorter section (2), focused on the effects of this approach on the logical, methodological and empirical levels of scientific research.
In the last few years some of the most prominent sociologists have highlighted in different ways 2 the fact that, with the decline of a reference system of shared values to build a person's life path, the individual is forced to rely on him/herself as the sole criterion for choice; even family bonds have lost all traditional references, and the forms they have assumed are justified by purely individual reasons; self-fulfilment has become the individual's primary goal and bonds are established for instrumental reasons -hence, the remarkable decline in their value and the idea that they belong to a distant past.The myth of romantic love is radicalised, as intimate relationships are mostly identified with the couple rather than the family; in this new scenario, the beauty of the self seems to lie within the couple.This cultural model is known as "individualisation".
In particular, Bauman (2003) highlights what he calls the liquefaction of bonds, that is, the tendency of personal relationships towards disaggregation and revocability, as both loneliness and romantic attachments can produce insecurity."Loneliness spawns insecurity -but relationship seems to do nothing else.In a relationship, you may feel as insecure as without it, or worse.Only the names you give your anxiety change" (Bauman 2003, p. 15).Beck (Beck and Beck-Gernsheim 1990) talks about an individualised marriage based on the partners' self-fulfilment; in a risk society, this process places individuals outside the social schemes which had configured life choices in Early Modernity (Carrà 1999, Carrà andSantoro 2006); Beck (Beck and Beck-Gernsheim 1990) defines the situation of today's couples as the normal chaos of love: a massive social overvaluation of love flooding the advertising and entertainment industries, the decline of shared social norms, and the isolation of the couple from kinship networks which lead individuals to keep inventing new rules and to evaluate their successes or failures by non-univocal criteria, thus making day-to-day love relationships remarkably hard to sustain.See: Giddens (1979Giddens ( , 1984Giddens ( , 1990Giddens ( , 1991Giddens ( , 1992Giddens ( , 1994Giddens ( , 1999)), Beck (1986Beck ( , 1996Beck ( , 1997Beck ( , 1998Beck ( , 1999Beck ( , 2009)), Beck and Beck-Gernsheim (1990, 2001, 2004), Baumann (1995, 1999, 2000, 2001a, 2001b, 2003, 2005).In fact, marriage and the family had been the target of criticism well before the time of Giddens, Beck, Beck-Gernsheim and Baumann.Theories on the decline of the family have existed since the beginning of the sociology of the family.Durkheim, for example, raised serious concerns about the family's capacity to perform important functions in the modern era and speculated upon a society in which professional groups would replace families as the main source of moral solidarity.In the 1920s and '30s, a large number of empirical surveys in the United States envisioned a historical phase of massive family disorganisation (Popenoe 1993).In the 1960s and '70s, some radical intellectuals and scholars actually talked about the death of the family, not so much in the name of individualism but intending to substitute the family with a better community, i.e., the commune.They saw the family as a depersonalising agent limiting individual freedom and thus needing to be eliminated; they did not intend to hail the birth of an independent, isolated individual but to urge a communality that would generate authentic relationships based on voluntary participation, which was considered impossible within the family.Giddens (1990Giddens ( , 1991Giddens ( , 1992Giddens ( , 1999) detects a considerable change in the meaning attributed to marriage: the basis of affective bonds no longer lies within reproduction or family-making but in a pure relationship, that is, one depending on emotional communication, where continuity is based on the advantages deriving from the relationship itself.This theoretical approach stresses the concepts of intimacy, self-construction and pure relationship, and postulates the central role of sexuality, the equalisation of homosexuality and heterosexuality, the interchangeability of gender roles, the transience of bonds and the continual reversibility of choices: it is a "democracy of emotions".
In the new millennium, however, an ever-increasing number of scholars have been sceptical about using individualisation as an interpretative criterion.The need for a fresh outlook was apparently prompted by longitudinal empirical studies witnessing the persistence of the family and marriage (Smart 2007, Morgan 1996, Finch and Mason 1993, 2000, Carsten 2004, Gillis 1996, 2004, Chapman and Hockey 1999, Miller 1998) versus cohabitation, divorce, and the spread of single-parent families; these phenomena, however, vary in distribution, visibility and social approval (Laslett et al. 1980).
Thus, in response to these unexpected data, a number of sociologists of the family have sketched a more complete and sophisticated conceptualisation of family life, whilst seeking new ways to capture its multidimensional relationships; this has urged them to work on the concept of relationality, and develop an approach antithetical to the dominant argument of individualisation.This common intention has produced a more proper relational trend and a relationist one (Donati 2011).Without delving into details, it is hoped that an illustration of the main features of either trend will cause the two different uses of the concept of social relationship to emerge by opposition (Rossi 2012).

The Relational Approach
From a sociological viewpoint, the development and dynamics of the family can be fully grasped with reference to theories on social morphogenesis.These, in fact, can help understand the differentiation processes at work in contemporary societies, such as the attempt to consolidate the living arrangements of some groups of people: although these may become settled for a time (morphostasis), they must always be screened according to certain criteria to decide whether a morphogenesis has taken place.The relational hypothesis (Donati 2006a) postulates a family genome, 3 that is, a latent structural pattern present since the beginning of the history of civilisation 3 A diametrically opposed position is that of Saraceno (2012), who studies the family from a constructionist perspective: she believes it is society that defines which couple and generation relationships are acceptable and recognised as family, and can thus acquire juridical relevance.These definitions have changed throughout history and in different cultures, just as the subjects have changed who are entitled to the right/duty to define the family, the obligations and responsibilities connected with family bonds, and the distinction between (or assimilation of) couple and family.On these historical differences are grafted, according to Saraceno, all changes produced by demographic and cultural processes.Population ageing, an increase in the number of unmarried couples, and homosexuals' demand to have their unions recognised are modifying both the idea of the couple and the processes of family formation.Finally, the possibilities offered by assisted reproductive technology have challenged the obviousness of the biological bond between those who generate and those who are generated.At the crossroads of demography, history, culture and norms, the family appears to be, according to Saraceno, a kaleidoscopic phenomenon. in all past cultures (namely, a cultural universal) and consisting of a dual relationship with unique characteristics: it connects the male and the female genders (see footnote n. 1), produces vertical bonds between the generations and creates interrelated genealogies.
Being permeated with an individualistic culture, contemporary man feels free to experiment with novel ways of articulating the family genome inherited from previous generations, thus producing a spectrum of different living arrangements.It would be incorrect, however, to regard this as a morphogenesis of the original pattern, that is, to think of a number of different family forms all equally falling within the family concept; it is rather the case of different lifestyles resembling, to a greater or lesser degree, a family pattern which, in fact, remains unaltered.
The concept of the family genome helps identify the family's DNA, the form (morphè) that has survived until the present day in a range of different interpretations through the family's various morphogenetic processes (Archer 1995).Viewed as a "total social phenomenon" (Lévi-Strauss 1967), the family, according to Donati (1986), "presents itself as a communality of undifferentiated aspects of daily life which are, nevertheless, economic, moral, political, religious, juridical, social, psychological" and is more narrowly conceptualised as an intrinsically relational phenomenon that is unique and irreplaceable -in other words, as a network of relationships.
In order to identify the family in this changing scenario it is necessary to sidestep a number of theses based on incorrect assumptions.As Donati suggests (2007), the positions to be avoided are exclusivism, which denies variability in the form of the family and is disconfirmed by the data; inclusivism, where any form of living arrangement is equated with the family; and an indifferent, neutralising pluralism, whereby the family is the form of daily life chosen according to a person's varying needs and preferences.
As anticipated above, when talking of the family genome, the family can be identified as a social relationship of full reciprocity between sexes and generations, to be neither replaced nor confused with anything else (Donati 2006b): "The latent structure attributing social identity to the family, that is, eliciting the unique social relationship we call a proper family relationship, consists of four interdependent elements: gift, reciprocity, generativity, and sexuality as conjugal love" (Donati 2006a, p. 58, translated by the author).
On a social level, the genetic endowment of the family is due to the coexistence and interdependence of four characteristics (Donati 2013): • Gift (to give first) • Reciprocity (symbolic exchange, mutual help) • Conjugal sexuality • Generativity (to have or at least desire offspring) It is opportune to explain the meaning of these four characteristics.
Gift.Within the family, one learns a dimension that is not acquired in any social context and that is "gratuitousness".Commercial logic does not exist within the family because, should it be introduced, the sense of family would cease to exist and it would lose its identity.As shown from studies of American social psychology (McCullough et al. 2001) the gift can be acknowledged with gratitude.
Reciprocity.The anticipation of reciprocity, unlike that of business contracts or financial transactions is only between those recognised as being part of family ties.What the rule for marital relationships and family relationships?That reciprocity is a symbolic exchange that can be deferrerred ("I give you something and I expect something in return").Even if this (mutual) exchange does not take place, the expectation remains.
Generativity.This is expressed by having children, but not only that.There is also social generativity in families, as shown by adoption.Having children, not as personal self-fulfilment, but as result of the common good within a couple.
Sexuality.And finally, sexuality, defined as conjugal love.Sexual intimacy, yes, but not a random form and not with random people.Spousal love is only for those who love with that kind of affection.
The very etymology of the term "relationship" thus evokes a dual series of meanings: first of all, the Latin re-ligo, ("to inter-link"), that is, to establish a bond between two or more subjects: this connection, or structure, inter-subjectivity, interaction can function as a restraint or a resource.Secondly, "relationship" relates to re-fero ("to refer to"), thus carrying a set of shared symbolic references, a sort of memory: it introduces the bond into history, thus linking it to other bonds that make it significant (Rossi 2003).
Thus, the family relationship presents a structural aspect and mutual expectations deriving from the bond; there is a (non-economic) exchange between the subjects, expressed by the concept of religo; furthermore, the subjects carry4 a cultural heritage which they represent within the bond.The couple is, inevitably, an encounter of two worlds, or two histories, that cannot be ignored, as each subject becoming part of a family is, in turn, a node in a generational plot (Scabini and Cigoli 2012).
A family relationship, however, cannot be reduced to either refero or religo: being both, it does not even amount to the added-up features of its constituent individuals.In fact, wherever a bond is formed, the subjects' history, or plot, is altered, thus generating novelty, a surplus which can only emerge when considering the relationship itself rather than just the individuals.Therefore, the family relationship should be viewed as an area enclosed within what Donati calls the referential, structural and generative semantics (2006a).The two aspects that are essential to a definition of the family, i.e., refero and religo, may actually oversimplify it; only the family's generative dimension can help assess the depth of the relationship and its capacity to generate something new, and shatter the two individualisms by planning a common action and a common future.
Furthermore, the internal dynamics of family relationships, particularly inter-generational ones, emerge with family transitions.A family transition is a critical phase following an event which significantly modifies a family's social system and requires an adaptation of the family's relationships, by the end of which a new equilibrium or, better, a new modus vivendi must be found to follow a morphogenesis of the we-relationship (Donati 2011).

Relationist/Relationistic Approaches
The relational approach is based on relational metaphysics and, therefore, on an ontology of relationships where these are regarded as the fabric of every society.The other current family studies perspectives that refer to the relationship category when dealing with family bonds do not adopt an ontological viewpoint when focusing on relationships: conversely, they take as their starting point the fact that relationships exist, and that they can be experienced and observed, just as their impact on individual's lives can be observed -these lives actually prove not to be individualised at all but deeply embedded in their relationships.This is why it is possible to justify the umbrella term relationism to cover these approaches while indicating their closeness to relationist philosophy, in which the sui generis nature of the social relationship is not postulated.
Since, unlike the relational approach, this is not a strictly sociological one, the relationism label can be applied to different positions all sharing the attempt to sidestep the definition of what a family is, in favour of a more comprehensive approach, focusing on family bonds as extensive, multiple, dynamic relationships.A number of scholars have adopted this multiple relationships approach to re-focus on the family (e.g., by the configurational approach) or give rise to specific research trends (e.g., life trajectories, intergenerational relationships, refugees and migration, etc.).
In more detail, it is possible to distinguish between relationist approaches in the strict sense and in a broad sense.The former includes Widmer's configurational perspective 5 which, far from being limited to a theoretical program, investigates family-making processes on the basis of empirical data (Jallinoja and Widmer 2011) in different constellations and through different transactions; these processes, in fact, can highlight some essential elements in the late-modern family.
Among the broader relationist approaches, a number of authors, such as Morgan (1996), Smart (2007), Finch andMason (1993, 2000), study the family according to a purely phenomenologicalconstructionist viewpoint, starting from daily life practices.
After a concise illustration of the main elements of these theories, the analogies with, and the differences from, the relational approach will be highlighted.

Widmer's Configurational Perspective
According to Widmer, the family must no longer be viewed as an unequivocally evident social unit: this viewpoint, in fact, obscures the mobility and changeability of family boundaries -an aspect that can effectively be highlighted by a concept of family formation as based on ongoing assembling processes.Widmer uses the term assembling to stress all the actions and transactions (gatherings) drawing together all those who, at any given time, belong to certain family.
According to Widmer, observing family formation as dictated by the "assembling processes" allows for a widening of the actual notion of family membership and redefines the boundaries between family and kinship, family of origin and family of choice, pre-and post-divorce families.With the help of this theoretical framework, Widmer arrives at conceptualising families as configurations, 6 that is, networks of interdependent individuals, characterised by cooperation, power and conflict; he maintains that such a vision, together with a network approach,7 could 5 For a more comprehensive discussion, see the volumes edited by Widmer and Jallinoja in 2008 and 2011.

6
The concept of configuration goes back to the works of Moreno on sociometry and sociodrama (Moreno 1934) and those of Elias on identity and community construction (Elias 1978).An interesting in-depth study on the configurational approach is Jallinoia's genealogical proximity ( 2009): she believes obituaries can help determine family configurations according to genealogical proximity.contribute to a better understanding of the complex family forms typical of Late Modernity.This outlook is based on four basic assumptions (Widmer and Jallinoja 2008): • families are not just defined by institutional criteria (such as marriage and cohabitation) but are based on actualised relationships; • dyads should be considered part of a wider network of relationships -that is, the conjugal bond is influenced by extended family relationships; • both the structural aspects of family configurations and the individuals' identities, perceptions and projects must be considered; • the historical and spatial dimensions are emphasised: all human configurations, in fact, evolve in time and space on the basis of social, cultural and economic changes. 8From a methodological viewpoint, the configurational perspective studies the various family experiences through the social networks, thus capturing the interviewees' perception of their family configurations: by responding to a series of stimulus questions, they must reconstruct their personal relationships and the entire web of interdependencies involving all the individuals in the network.This methodological setup avoids the critical node of an empirical research based on a limited number of family dyads, made irrelevant by the complexity of the contemporary family contexts; conversely, the whole set of overlapping relationships present in each configuration must be considered.This approach also privileges longitudinal studies aimed at investigating the evolution of interdependence bonds.

Practice-based analysis: Morgan, Smart, Finch, Mason
The scholars presented below share an outlook based on the daily practices of the family and aimed at capturing the multidimensional experiences that define it.They are members of the Morgan Centre for the Study of Relationships and Personal Life, a research institute of Manchester University, founded in 2005 with a focus on family issues, relationships in Post-Modernity, parenthood and childhood. 9The Centre is aimed at developing new theoretical perspectives on personal life and intimacy, besides pursuing new directions in empirical research.

Morgan's Family Practices
In his book Family Connections, Morgan defines the family as an aspect of social life, not a social institution (Morgan 1996, p. 186).He proposes the expression family practices, to capture the multiple dimensions of family life and, particularly, to: • record the observer's and actor's viewpoint; • refer to the action (unlike the static family structures); • focus on daily life and give actions a sense of regularity and fluidity; • connect the historical dimension with the biographical one.
others" (i.e., anyone playing, or having played, a positive or negative role in the subject's life), the interviewee identifies existing relationships, and is then asked about the emotional support, conflict or influence derived from each.This method highlights the interviewees' perception of the structure of the family configurations they belong to. 8 Strictly connected with the theme of the plurality of family forms and experiences as analysed through the configurational approach is Widmer's reflection on "Family or Life Trajectories", that is, the multiple, variegated, family life paths.The concept of trajectories intends to go beyond the idea of the lifecycle (which postulates a set of definite, specific phases following one another in a person's life), in an attempt to capture the uniqueness of each path, by understanding it as a specific form of family diversity.
His analysis is based on the actions characterising the different aspects of family life.This shifts the focus from the family as a structure to which individuals simply belong and which, in a sense, precedes them, to the family as a set of activities that assume a particular meaning, at a given point in time, through the practices shared by its members.

Smart: The Toolbox of Concepts; New Directions in Sociology
C. Smart theorises a toolbox of concepts, that is, a set of instruments which, however limited, can be used for producing a collage of family life by which to access its hidden texture (Smart 2007).Relationality, Memory, Biography, Embeddedness, and Imaginary can be seen as overlapping areas for theoretical exploration.

Relationality
The term relationality indicates that persons are always connected to others who are not necessarily related to them by either blood or marriage; thus, people exist within intentionally created networks in which a great deal of intellectual and affective investment is made and which they support, maintain, and/or eventually neglect.This concept prioritises the consideration of family of choice over family of origin (Weston 1991).The combination of relationality with practices (Morgan 1996) challenges the idea of relationships as unalterable because genealogically given.However, it must be noted that an emphasis on relationality (Mason 2004) does not per se connote relationships as positive but refers to their influence on the daily choices of individuals.

Memory
Our first memories are formed within the family context and take on special meanings due to the nature of early experiences and due to the fact that they concern persons who are important to a young child (Mauthner 2005).Families are mnemonic communities that influence and shape what and how to remember; by creating shared memories, they give their members a common history and therefore an identity (Misztal 2003); this sharing produces familiarity and strong bonds, beyond any positive or negative connotations that individual memories may have.

Biography
In order to attain a deeper knowledge of relationships, a biographical turning point is necessary (Rustin 2000).This implies acknowledging the explanatory power of the case study approach to give a voice to people's motivations, desires and aspirations and perceive the meanings that individuals attribute to events and relationships.

Embeddedness
Some qualitative empirical studies conducted from the mid-1990s (Finch and Mason 2000, Mitchell and Green 2002, Smart 2005, Brannen, Moss and Mooney 2004, Bengtson 2001) have highlighted how the life of each person is embedded in a web of past and present relationships.Embeddedness per se has no positive (or negative) connotations, in that it only identifies the strength and persistence of the relationship, that is, the extent to which people feel bound as being part of a whole.

Imaginary
Imaginary defines how relationships exist in our imagination and thoughts; it is not limited to one's projecting ability but is formed and influenced in a historical, cultural and social context.The imaginary family takes the form of expectations and desires linked to minor or major family events -i.e., planning holidays, organising family gatherings, collecting smiling photos for the family album, banning all talk of illness and death -and is constructed through family practices.

Finch and Mason: Kinship Networks and Generations
In "Passing on.Kinship and Inheritance in England", Finch and Mason expound their empirical analysis of the transmission of inheritance following the death of a family member. 10To study how inheritance is managed contributes to the definition of the blood/marriage relationship and its rules.When making a will, one needs to choose who, among one's closest relatives and acquaintances, may be important enough to receive part of one's property.Inheritance, which may consist of both material and symbolic goods, is viewed as a process contributing to moulding (not just reflecting) the idea and meaning of the family.Finch and Mason maintain that a main feature of the kinship network is the high level of flexibility by which individuals choose who to insert at which level.In fact, no universal rule exists: the only trait liable to generalisation is perhaps the innermost placement of biological parents and children.Thus, kinship appears as a set of flexible, variable relational practices: rather than being a structure that precedes individuals, or which they access through specific events, it is actively constructed by individuals through practices that are relational in time and space.

Assessing the Approaches
The above approaches have certainly contributed to overcoming the individualisation perspective through the concept of relationality; their attention to the way relationality is articulated in daily life is also significant.There are, however, some notable differences with the relational approach.These are detailed as follows: • the concept of assembling evokes the image of a construction in which blocks can be added and mixed at will in the absence of a solid foundation; so, the sense of the morphogenetic process is lost11 together with any references to a changing latent structure; • hence the scarce relevance given to gender (see footnote n. 1) differentiation which, conversely, is substantial to • the relational approach; • so it seems that these relationist outlooks cannot avoid either inclusivism or an indifferent and neutralising pluralism, both of which are based on incorrect assumptions; furthermore, according to relational sociologists, they confuse the family pattern with what simply resembles it; • this seems connected with a difficulty, or a reticence, about conceptualising the bonds' ethical (refero) dimension: rather than implying a threefold order of meanings, the family relationship is confined to the re-ligo axis, that is, to mere interaction, with weak, purely affective, connotations, displaying indifference to both genders and generations; • finally, this concept risks over-focusing on the synchronic dimension at the expense of the non-diachronic one, making it impossible to fully grasp the relational surplus, the family's proprium, the depth of the relationship as given by the generative semantics.With an exaggeration, it could be said that the theories considered struggle to grasp the essence of the family as a social relationship that only exists and is possible within time -or in a relational time connoted by a historical mode.The time frame they refer to is rather an interactive one (factual mode).This has two basic consequences: • difficulty in focusing on family transitions, resolved by the term trajectories as an alternative conceptualisation evoking the multiplicity, subjectivity and variety of the family's life-paths; • a weakened family, intended as a more or less elusive network.Consistently with this last point, the method used cannot be defined as fully relational: in fact, its specific object is the individual in his/her relationships, not the actual relationships.Despite the number and diversity of the approaches and instruments adopted,12 there is a latent risk to arrive at an understanding of the family that is one-dimensional or, at most, two-dimensional (present and past).If, as highlighted by Jallinoja (Jallinoja and Widmer 2011), Finch andMason (1993, 2000), there are privileged times for observing the family (e.g., death, dealing with inheritance, etc.), there are two methodological risks: • equating the nature and identity of the family with the situation of a given family in given circumstances -where reflexivity may be scarce or absent and the family might be undergoing a process for disorganization; • confusing the analytical and institutional aspects (e.g., by regarding inheritance as a family-making process rather than one that simply reflects families).

Between Individualism and Relationism: the Interpretation of Relationality in French Sociology
Despite their differences, the above approaches have recently converged on the urge to go beyond individualisation by recovering and valuing the concept of relationality.This attitude is shared by other authors, who hold different views on the individualisation process: in France, family sociologists have devoted their attention to the transformation of family bonds, offering a number of interesting solutions.
François De Singly talks about an individualisation process (1996): the fulcrum of family relationships is no longer the couple, the we, but the partners' identities, as they use their family relationship as an identity-generating instrument and a means to self-fulfilment.Interestingly, De Singly does not think this process spreads despite the family; on the contrary, individualisation has become the cornerstone of contemporary family theory (Cicchelli Pugeault and Cicchelli 1998, p. 108-109).In his hypothesis, the family remains crucial to a "self-revealing" process, since an individualised contemporary identity actually requires "a certain type of relationship with others" for its self-definition (De Singly 1996, p. 13).In "Libres ensemble" (2000), De Singly effectively outlines some salient aspects of the link between individualisation and family: • the person is considered a subject from birth, it is him/herself before being "the child of"; • the conjugal relationship is prioritised over those with the families of origin, from which the duty of love derives (the relationship is based on love rather than on a pact between families); • the relationship becomes an instrument for self-knowledge: so, if one proves to be different from the way one used to be, the relationship should "naturally" be questioned or even interrupted and replaced with another, as no longer fitting one's real "I" and if the couple cannot tolerate the inevitable compromises that protect the union.Kaufmann, on the other hand, sees strengthening of the couple and exaltation of the ego as complementary; through an original analysis of daily life (1992,1993,1999,2002,2007), he highlights the ambivalence of the relationship between the we and the I.
This approach is taken further by Théry, who uses the term démariage (1998, p. 32) to indicate the progressive de-institutionalisation of the man-woman relationship, no longer constrained within a recognised, unconditional, conjugal bond: the couple relationship is continually renegotiable.Théry, however, stresses the reticularity of family experiences and the need to observe them longitudinally: she believes that today families are to be defined in terms of the parentchild bond, contrary to the traditional and modern idea of the family as based on the couple relationship (which is currently posing a number of political and social problems).So the focus of the generational history is no longer the conjugal pact but the non-negotiable filiation pact (Théry 1998, p. 34), whereas the couple relationship is subject to continual bargaining: where marriage is de-institutionalised (démariage), the parent-child relationship becomes the new social institution (undergoing an ever intensifying legitimation process).Besides, filial relationships too undergo transformation, as the overlapping of children and parents prevails over their succession (because of increased longevity, one can simultaneously be both a parent and a child within the same family), with a child-to-parent inversion involving intense exchanges.Family identity can thus settle as generational identity (of a generated and generating subject) whose bonds are nonnegotiable.

Some Critical Remarks
The attempt to keep family and individualisation together is shared by some eminent sociologists from the French school (De Singly, Kaufmann, Théry), although their approaches are differently positioned between the two poles of the family-individual relationship.Without going into details, it can be pointed out that a focus on the individualisation process in order to revitalise the reflection on the family is a risky process, as the boundary between individualisation -a tendency towards self-autonomy -and individualism is extremely blurred.
In addition, the real capacity of individualisation to produce a generative couple needs questioning.In fact, the features observed by these French authors do not exhaust the essence of the couple, which remains a relationship within which the partners' relational networks converge directly or indirectly, evidently or tacitly.

The Added Value of the Relational Method
In the setting of a scientific research project three levels can be distinguished: logical, methodological and empirical.The logical level concerns the theories and hypotheses on a certain object of investigation; the methodological level refers to choosing, and dealing with, the relevant information; its function is to mediate between the theory the hypotheses are based on and the empirical level (i.e., the tools and techniques chosen for information gathering and data processing).These three levels actually overlap, and the transitions between one and the next follows a circular path.The quality of the overall research is given not just by that of each level but also by their compatibility and consistency.For this to happen, the reference theory and the identity of the object of study (with its specific traits) need to be defined clearly (Iafrate 2003,

The family as a unicum
Not just the sum of its members

Non-independence of family data
The persons belonging to the same group are more similar among them and have more similar perceptions than those belonging to different groups Lanz et al. 2015).On the object of study and the specific traits to be considered, a high degree of problematicity emerges in connection with the method of scientific research applied to the family (Donati and Scabini 1990, Lanz and Rosnati 2001, 2002) (Tab.1).
The first specific trait is the fact that, in the case of the family, the unit of analysis is a unicum,13 and that the researcher needs to obtain information about both the family and its components.«The family is a unitas multiplex, a complex system that must meet the challenge of maintaining both its unity (as a whole) as well as giving space to relationships among single members.This means that the whole and the parts have to be simultaneously taken into account.The idea of the family as a complex unity is central to the relational-symbolic model of family research and intervention (Cigoli and Scabini 2006), which particularly develops the intergenerational aspect of family bonds» (Lanz et al. 2015, p. 2).
The second specific trait is linked to the relational and historical nature of the object family: the fact that a family owns rules, roles and a history affecting the relationships within it must be taken into account by the researcher.
Another specific trait is the non-independence of family data, based on three factors: the family's compositional effect, a common destiny and a reciprocal influence (Kenny and Judd 1986).Composition means that family members and their roles are not assigned randomly and are not interchangeable; their common destiny is determined by their sharing the living environment and a set of expectations for the future; finally, relationships among members of the same family are characterised by mutual influence.
In short, research on the family is intrinsically complicated by the nature of the relationships within it (Bray, Maxwell and Cole 1995).Such complexity engages today's scholars in their attempt to devise compatible methods and techniques (Tagliabue and Lanz 2010).
For all the family researcher there are a number of methodological options (Tab.2; Gonzalez and Griffin 1996): none of them can be taken for granted or treated superficially, given that what is known is inseparable from the way it is known.Transversal/ longitudinal studies However, most of the studies carried out so far are affected by the difficulty of shifting the focus from individual members to the family as a unicum.In this context, the relational approach does not reduce its epistemological and cognitive tension: rather than family-related studies, it aims to produce family research projects, where the logical, methodological and empirical reference level is the family.
There isn't a very distinct separation between family research and family related study.It's possible to outline a continuum between a minimum (family related studies) and a maximum level of rationality (family research) (Fig. 2).
According to relational sociology, this aim can be reached through an established methodological instrument, the AGIL scheme 14 , which helps to analyse the family as a sui generis social relationship, by identifying its constitutive resources (A), aims (G), norms (I) and latent cultural model (L).This identifies the subjective and objective elements (respectively: 1., value orientations and attitudes, and 2., adaptation conditions, i.e., means and norms independent of the subject's will) that generate the family as a social fact.Furthermore, it is necessary to consider both the internal viewpoints of the family relationship as it is viewed (carried out) by its subjects and the external social relationship.
The different family forms can be described according to the way they combine the four fundamental aspects of the genome.If we apply AGIL to the family, on A we can find sexuality as a means to join male and female genders within a conjugal relation, on I reciprocity as the rule of the relationships between genders and among generations, on L the gift and trust code that steers family relations and on G generativity in all its forms as the goal and the outcome of the family relation.We achieve the aim to understand whether, how and to what extent the requirements of the analytical framework are present, how they relate each other, how they influence each other, and finally which configuration the observed relationship actually has.In this sense, contemporary living arrangements can be analysed through AGIL compass, in order to measure their closeness or distance from family genome.
Basically, the above analysis -which can be defined as relational -is, first of all, the toolbox to be used in the study of social phenomena from a relational viewpoint; in fact, besides providing instruments, it contains a methodological map indicating the sociologist's correct attitude when using it, in connection with the idea of avalutativity.Addressing the theme of avalutativity of relational analysis, Donati first of all clarifies that "relational sociology is a positive science but not positivistic" (Donati 2006c, p. 234, translated by the author), because as proven, it does not reject any significant datum based on judgment of value, at the moment in which the analysis takes place, "but it does not ignore that the material being processed is 'strongly of value'" 14 See Donati (2011, p.153) «Relational sociology conceives of the AGIL scheme as an analytic tool that retains the four "poles" A, G, I, L, (as they were obtained by Parsons), but it interprets them as "orientations of meaning" attributed to these elements (constitutive of social phenomena) by the agents/actors who generate and sustain them.
[…] Relational sociology sees the components of AGIL as being relationally generated and in turn affecting social relations as such.Relational AGIL describes the form of social relations as a sui generis order of reality (having its own properties and causal powers) that emerges from reciprocal actions (which, in their turn, have their own and different AGILs)».(Donati 2006c, p. 234, translated by the author) and, on the contrary «the observer must relate the facts to his/her own ethical values, explaining the relation according to the ethics to which reference is being made» (Donati 2006c, p. 235, translated by the author).
In addition, the relational approach can be an interesting and profitable interface between psychologists and sociologists.Family research orientation and interdisciplinarity are the distinctive features of the studies carried out by the Centro di Ateneo Studi e Ricerche sulla Famiglia15 in partnership with other Centres or research Institutes.
Therefore, the study of the family requires empirical research projects that may avoid oversimplification or a loss of the family's relational specificity (Larsen and Olson 1990).The relational connotation must affect all the stages of the research, to avoid a gap between the theory and the methodology adopted. 2

Fig. 2 .
Fig. 2. The Continuum between family-related studies and family research