Relational Social Work


Vol. 3, n. 1, April 2019

(pp. 2-14)

doi: 10.14605/RSW311901

The relational gaze in Social work. Four vertical levels of professional observation

Fabio Folgheraiter

Catholic University of Milan, Italy


Fabio Folgheraiter



The article is a theoretical reflection on professional observation in social work. Practitioners which deal with social problem must take account of the complexity of the system that observes those social problems and should be aware that they are part of the system. The paper invites social workers to take account of three levels of social observation: the level one is related to distinct observers seeing an external physical reality or a problem correlated to it; the level two involves some observers seeing a de facto reality which includes other observers of it and the level three is related to some observers which see a reality which compromises possible observers of further observers. To ascertain the range of possible perception is a social workers’ task in every situation. Moreover social workers must see themselves as they observe the entire observing system arrayed below them. This entails that social helping work has the formal structure of supervision.


Professional Observation – Social Work – Social Problem – Observer System

The most powerful taken-for-granted assumption in social work practices is individualism. «Thinking in terms of individuals» means conceiving the events that make up the problem as a set of items which gain unitariness and meaning in the identity of an individual. This mode of thought forces the narrator to represent the problem as if it resides in isolated individuals, who may be him/herself or others. It also induces the narrator to identify the causes of the problem as being in individuals, again him/herself or others. This unconscious psychologism gives rise to the appropriation of further questionable explanatory codes based on the concept of «pathology», codes drawn from the medical domain, and which the latter in its turn has appropriated (somewhat haphazardly) from common sense.

We term «medical» the mode of thought according to which the individual whose living is viewed as «a problem» is such because s/he is «sick». And it is the distinctive belief of the medical domain that the solution of a problem is always the healing of the individual affected, his/her liberation from the disease, even if this is not possible at the moment. For example, in the case of a person who appears «disabled», the medical code has us believe that it is natural — the only sensible thing to do — to focus on the «handicap» and to eliminate it. If technology permits it, this operation is performed immediately. If it does not, we will long for that technology, hoping we will be able to do it tomorrow or in a hundred years. It is not the material repair of illness that connotes the medical domain: rather, it is the mind focused on this aim and the belief that this operation is the only thing possible or conceivable.

The intuitive assumptions of the medical model are so strong that even an expert social worker may fall into the trap1 of accepting these apparently obvious premises as commonsensical, stretching the idea of healing out of the organic area and applying it to any kind of mending, even where medicine has no power. This medicalistic drift poses serious professional problems to the social worker. Strictly speaking, he should sift out covert distortions that he listens from his interlocutors with his cultural filters, which by definition are non-medical. However, a social worker may easily assimilate «erroneous» commonplaces and adopt the medical model wholesale. This happens partly because he is a human being just as anyone else, who belongs to the same culture as his interlocutors and shares their intuitive assumptions, which his training has been unable to eliminate. Academic curricula in social work should dispute common-sense assumptions and demonstrate their deficiencies. But in reality they accept them, and sometimes even endorse them. The majority of the so called theoretical models used by social work — psychoanalysis, behaviourism, the systemic approach, for example — are shaped by «medical» determinism, perhaps because of their desire to emulate a more prestigious science and profession which has constructed fortunes behind the glamour of the white coat. But when one reasons with logical shorn of any misleading unconscious interest, it is evident that this tendency gives rise to numerous risks, as well as patent absurdities. Social work seems allergic to medical and paramedical determinism; if it assimilates too much, it is poisoned and dies.

Social observation and the stratification of observer systems

A practitioner who deals with social problems, rather than with broken or malfunctioning biological structures, must constantly take account of the complexity of the systems that observe those problems. A problem is said to be social when it arises from the numerous observers present in its setting, ranging between those most closely involved in it (users and family members, for example) and those more detached. The social worker is one of these observer systems (Folgheraiter, 2004). Because of his/her role, s/he cannot ignore (indeed, must pay particularly close attention to) a possible absence of observers, or even to some sort of selective blindness which produces a lack of relevant observations.

Figure 1.1 illustrates the possible interweaving of observer structures present in a hypothetical social problem. It shows various observers at different levels who see different or the same entities or phenomena, which also include themselves.

Within the small circle, in the first level, three observers in direct contact with an external «phenomenon» (the rectangle) are represented. We may imagine, for example, that they are three doctors inspecting an X-ray which shows a spot on the left lung; or that they are a psychologist, a social worker and a gynaecologist who «observe» a pregnant fourteen-year-old girl during distinct interviews at a child and family centre.

Whatever the reality observed, it is assumed to be what its name says it is: namely «real» like something really existing in nature. I disregard for simplicity the more refined theoretical point that every phenomenon, even the most concrete ones like rocks or trees, can be conceived as constructed by the observer and therefore as non-real beyond their outward appearance. The observer uses his/her mental constructs to perceive and define the phenomenon, and therefore sees it according to its distinctive features.2 For instance, if someone is walking along the street and sees «something long and thin on the ground», s/he may see a stick or a cord or a snake according to how his/her mind codifies the object’s features and attributes meaning to them. S/he may see a snake (and scream) even if closer inspection shows that it is only an innocuous stick. But if it actually were a snake, it is evident that this specific label for the long thin object on the ground derives from the concept of «snake» in the observer’s mind (and perhaps from an instinct inherited from natural selection).

Level 1 of social observation: distinct observers see an external physical reality, or a problem correlated to it

Given that we are obliged to engage in abstract analysis requiring firm intuitive bases, we opt for a «realist» philosophy (Bhaskar, 1978). Confident that nature is not the amusement of an illusionist playing games with us, and that our perceptions have solid correspondences in reality and are generally trustworthy, we may define as «real» or as a «fact» any thing or event which produces stimuli activating a perceptive code, and therefore the construction of meanings that approximate the «real» thing-in-itself as closely as possible.

Fig. 1.1 Abstract observation levels starting from an objective fact.

Simplifying, we consider here to be «de facto reality» also a second-level perceptive coding: that is, a deduction or inference drawn by the observer from primary facts. A «problem» is essentially a deduction. As explained in more detail in the next chapter, the physical fact exists on one level, the problem connected with it on another. A spot on the X-ray is a de facto reality which the doctor must first perceive as a spot and then, having ascertained that it is one — perhaps also by asking a colleague for his/her opinion — must give it its proper meaning in terms of a problem, e.g. a cancer. Something long and thin on the ground is a de facto reality which the observer first perceives as a snake, rather than as a harmless piece of wood, and then deduces — in this case unreflexively — that s/he must avoid it in order not to be bitten. The snake is a problem by virtue of the observer’s concept of «snake» (a long, thin creeping reptile able to inject venom), from which s/he immediately infers the problem and how to avoid it. A small child bereft of this concept is unable to see the problem, although s/he sees the long thin object on the ground better than can any adult in the vicinity.

Once again, although «fact» and «problem» (judgement on the fact) lie at distinct levels, for convenience they are here considered to be interwoven. If a fact is taken to be real, so too is a problem connected with it. In the case of a doctor examining a chest X-ray, on seeing the spot s/he may deduce that it is a tumour almost by reflex, thanks to his/long experience of this kind of observation. Seeing the spot is seeing the tumour ipso facto: the fact is the problem. If a cyclist has fallen and has scraped his leg, we say that the sudden change in the state of his leg following the fall is a «problem» simultaneously with being a «fact». We have among our mental constructs the notion that it is normal for a cyclist to be on the saddle and for his leg not to be scraped, so that when we see the cyclist on the ground with his leg bleeding, we perceive the phenomenon as a problem that indubitably exists.

We assume that the phenomenon (fact/problem) exists in nature and we anchor its objective substance in its immediate perceivability. This does not mean that the human perception of this reality is unequivocal. A physical thing that exists in a certain place at a certain time regardless of any observer, so that even if no one perceives it we say that it still exists (although in this case it is difficult to conceive it without fantasising) has the property that, while someone sees it in a certain way, someone else may see it differently, and someone else may not see it at all, and so on. The «existing» reality is that which it is, but an array of minds may reproduce it in different ways,3 so that we know that it exists but we never get to the certain truth of whatever it is without the pleasure or displeasure (according to the case) of argument and debate.

The decision to merge fact and problem together produces further uncertainty, because the more reality becomes «jumbled», the more the reasons for divergence increase. The doctors may disagree over whether there is a spot on the lung (if the X-ray is not as sharply defined as it should be). If they agree that there is one, they may disagree on whether it has a bearing on the diagnosis (it may be due to a flaw in the printer). If they think that the spot is the sign of something in the lung, they may disagree on whether it is malignant or benign, and so on.

From each individual point of view, people are certain that the reality is what they see, but there are others equally convinced that what they see is true. To a greater or lesser extent, this assumption applies to every perceived phenomenon. Whilst perceptive divergence is possible in health care, it is ten times more likely in social work, owing to the greater mutability and complexity of social signs and the deductions made therefrom.

As will be made clear in the rest of this book, a social worker cannot be called such unless s/he is aware that social perceptions of a phenomenon are by their nature different (or differ in their intensity). Above all s/he cannot be called a social worker if s/he assumes that what s/he sees is what others see, or what they should see. The rest of the book will also show that this methodological imperative of recognizing and tolerating differences should not inextricable entangle the social worker in a web of relativism. For the purpose of social work, as we shall see, is to grasp similarities and help them merge together amid diversity towards good.

Given these assumptions, we must now investigate how the same fact/problem is perceived by different observers, and how different or similar perceptions and their correlated meanings may arise from the same external signals that present themselves to the senses of different human beings. I shall now discuss the possible logical combinations between the same objective «de facto reality» and the direct observers of that reality, providing further examples drawn from social work.

Different observers see the same de facto reality

Complete agreement among observers may arise in concrete and normal situations where perceptions are common and consistent. The same mental constructs are activated in unison, and each observer sees largely what the others see.

For example, an elderly woman is found by her daughter lying on the floor next to her bed with a cut on her head and in a state of confusion. The daughter calls a neighbour and her brother, who works nearby.

In this scenario there are the three observers envisaged by the first level in Figure 1.1. They all see the same thing (the elderly woman lying on the floor) and have no difficulty in agreeing. Indeed, the problem of agreement does not even arise because what is before their eyes is, they say, obvious. It may be that, although confused, the elderly woman is aware of what has happened. In this case, there is a further observer, in that the elderly woman is able to «see herself», so to speak, because she is conscious.

Some observers see one reality, others see another, yet others see nothing (even though they are exposed to it)

It may instead be that the elderly woman is unaware of what has happened, and is unable to see her head wound even with the help of a mirror. In this case, the woman, though herself a «subjectivity», presents herself as some other physical datum of reality to the sight of the others.4 She is not an observer of herself: she sees nothing about herself which the others are instead able to see.

Another example might be the following.

Two teachers are teaching a class in which a child has bruises on his arms and body. One of the teachers sees the bruises, the other does not, because the boy’s clothes cover his arms and body.

One may imagine that, if the first teacher points the boy’s condition out to the second, she will be able to see it on looking more closely. But it is more difficult to perceive a fact that lies at a different level of abstraction: for instance, the fact that this is not the first time that the child has come to school in this condition. Bruises appear on his body with a certain frequency. Let us assume that the less observant teacher has missed this sequence, which as such is no longer visible. Only if it is stored in the memory can it serve as the substrate for the current datum and generate a different, perhaps more severe, perception of the reality, to the point that a more complex deduction is made.

Some observers see a problem by deducing it from the fact, others do not (or see a different problem)

The teachers now aware of the frequency with which the fact «bruises» occurs may come to suspect, or even be certain, that the problem is more serious than mere dark marks on the skin: a problem of abuse in the child’s household. One of the teachers may be more certain about this «diagnosis» while the other may be more doubtful. One of the two teachers may also be more certain of the opposite and argue, for example, that it is a matter of pure chance, that the child has fallen over and hurt himself, etc. In support of this inference from the facts she may cite another set of facts, for example that the child’s parents are well-off and decent (of this she is certain). The deduction that the teacher draws from this fact (the family is fine so the abuse does not exist) is not logical but it may be accurate. But what matters here is that this deduction contrasts with the one made by the first teacher on the basis of the same perceived signals.

To return to the case of the elderly woman who has fallen at home, to agree that the fall and the wound exist is one thing; to decide whether or not these indubitable facts constitute a problem is quite another. Someone may argue that the incident is unimportant, that it is not even necessary to call a doctor, that once the woman is back on her feet, the fall is no longer a problem. Someone else may view the incident as serious and argue that the woman needs treatment and supervision. Even if all the observers agree that the medical problem exists and that at least a doctor should be called, another kind of deduction may be more difficult, a deduction that may be drawn from the former but that is related to differences of opinion. In effect, one of the observers may deduce from the elderly woman’s condition that, once the medical problem has been solved and the woman has been reassured, she will no longer be able to live on her own. This fact/problem different from the mere medical datum — the disappearance of the woman’s self-sufficiency — is of a social nature and is debatable. It may be clear to all the observers that the woman cannot live on her own; that is to say, they all perceive the risk that she runs in her condition. Or it may be that they all perceive it except for the son, who is particularly attached to his mother and does not want to perceive her situation as more serious than it actually is. The woman may not realize that she is at risk. She is unable or unwilling to admit that the problem of her self-sufficiency exists. She may therefore thank those present for what they have done, but tell them to go home and leave her in peace.

Level 2 of social observation: some observers see a de facto reality which includes other observers of it

At the second level in Figure 1.1, the second oval comprises three (n in abstract) observers in line with the same point of observation. They can see the previous level in its entirety, that is, the observers of Level 1 together with the de facto reality perceived by them. These second-level observers can see the objective reality (the rectangle) in the same way as the first-level observers do, and can accordingly be included among the latter. But they also see incorporated in that de facto reality a very different one: that the first-level reality possesses a derived reality which observes it. The first-level observers themselves constitute a fact apparent to the second-level observers; and this fact — the array of observers in the act of observing — integrates with the concrete reality (the primary fact) that stimulates that observation until it becomes wholly part of it.

A distinction should be drawn between a mere secondary observation which grasps only the overt action of the observing others and a more penetrating observation which grasps what is perceived and also, if there are several observers (as in Figure 1.1), the degree of their agreement or disagreement both on the bare facts that they perceive and on the problems that they deduce from those facts.

More prolonged observation may be necessary to understand this social aspect of observing and be able to say with some certainty whether the observing minds perceive the fact in the same way or differently. It may also be necessary to watch the observers’ interactions, listen to what they say, and see how they inter-relate in order to determine their perception and what they infer from it. This more profound observation is able not only to register «who» observes but also what (things and problems) the various observers really identify.

The second-level observers may be detached from the scene that they are observing, so that they see without being seen. They may, for example, be watching from behind a one-way mirror — as clinical supervisors do on certain occasions — or they may be standing at an apartment window and watching events in the building opposite. However, it more frequently happens that the secondary observers, too, are immersed in the scene and therefore, as said, physically adopt the same standpoint as the persons that they observe observing directly. Thus they too have access to the primary reality. In order to grasp this curious situation we must get used to thinking ambivalently — something that, incidentally, is typical of social work. We must imagine the observers as bi-focusing their minds and occupying several observation levels simultaneously. In order to perform this contorted twofold role, the mind must withdraw and observe from an external standpoint. I alluded to this logical schema in the example considered earlier, when I suggested that the elderly women who had fallen could be both the fact observed (the fallen, bloodstained body) and one of the observers of that fact.

On this reasoning, we may hypothesise that the elderly woman also stands at a higher level of observation. She may perceive and process with sufficient reflexive vigour the presence of the observers who have come to help her after the fall. Besides seeing and being aware of her fall, the woman may also be aware of those who can see its effects; namely, as said, herself, her two children, and the neighbour. The woman may be more interested in this observation than are the other persons present for affective reasons, because it demonstrates to her the affection and concern of the persons who have hurried to help her. The latter, too, are likely to perform this superordinate observation. Each of them may realize that others have hastened to help the woman, and they know who they are. But it is equally likely that, preoccupied with observing the crude fact, together with the possible «problems» connected with it, and therefore with how to solve them, they «look at» the other people present but do not «see» them. They do not process them reflexively, and they do not take cognizance of them intentionally. The elderly lady may be the most intent on scanning the people present because she is gratified by their concern and grateful to them. She also wants to «photograph» them because she believes in good manners and intends subsequently to express her thanks and return the favour. Moreover, the woman may carefully observe who is not there, noting for example that the neighbour who lives in the flat underneath hers, who was certainly at home and heard the noise of her accident, did not come upstairs to help her.

It is one thing to focus on the observers and fix them in the memory; it is another to determine what they have grasped by observing. If we take as «fact» not the simple fall, of which the woman herself is aware, but the broader problem of her non self-sufficiency, we find that others from the woman herself (assuming that she does not see the evidence or rejects it) are second-level observers of this phenomenon of obvious social-welfare concern. For example, the neighbour, besides seeing the problem on her own account, is also watching how the others react. She thus positions herself at a different level from them (unless someone else is doing the same thing). She may notice, for example, that the daughter — but not the son — has realized that it is now risky to leave the mother on her own. It may also be clear to the neighbour that the perception of the secondary deduced fact (the presence of a risk) excludes the elderly woman — who, as said, does not perceive it. Or it may also be that the neighbour is able (only her) to perceive (thanks to a long-standing relationship with the woman who has had the accident) that, despite appearances to the contrary, the elderly lady too is an «observer» of the risk: deep down, she is aware of the risk, but she refuses to acknowledge it, doing so either to repress an unpleasant reality, or to deflect a hasty decision by her children, or even to release her helpers from further obligations.

This double level of observation once again comprises the problem of agreement or disagreement among the observers. If there are two or more of them able to see the direct observers, they may perceive different realities. Once again, the risk of discordant perceptions concerns not who is present in the situation, about which there is little doubt, but «who perceives what». For example, if the observers of the perceivers of the woman’s «non self-sufficiency» problem are not only the neighbour but also the daughter, with a little luck we have the same «photograph» in both of them, but we may also have different ones. The daughter may count only the neighbour among those aware of the problem, while the neighbour may include the daughter and her mother. Then each of them may also see themselves. As observers of those able to see the non-self-sufficiency problem, they too must necessarily see it; otherwise they would not have in their minds the constructs indispensable for forming the category observed. But the question is whether one of them is able to «see herself» while she observes the other observers, or whether she is able to see the observation being performed in her regard by the other person occupying the second level. Should this observation occur, it moves the person to the next level of observation, which is more sophisticated but less abstruse that it might seem: Level 3 in Figure 1.1.

Level 3 and those that follow of social observation: some observers see a reality which comprises possible observers of further observers

The third level of observation can be occupied by any person who, in a contingent situation, is a first- or second-level observer, like the neighbour or the daughter in the example. But it is much more likely that it is occupied by a person whose job it is to observe: for instance, a professional social worker. Also a health worker could in theory occupy this level; but s/he would do so in an entirely accidental manner to do more with his perspicacity as a person than to his/her abilities as a professional. To resume the example:

Let us assume that the persons present agree that the elderly woman’s fall has given rise to a medical problem, or at least to a serious need to prevent such a problem, and have called the doctor.

The doctor arrives after a while, and he too, perhaps without realizing it, joins the array of observers already present. What does the doctor see? Or, put otherwise, what level of observation does he join?

He may focus only on the elderly lady; or indeed, out of professional habit or a sense of duty, he may concentrate solely on her symptoms and injuries. He is obliged to adopt this first level of perception centred on the facts alone by his professionalism, which requires him to be concrete and objective. It is therefore certain that any serious doctor will be an excellent level-one observer. We may consequently expect that, for example, he will notice a further bruise on the woman’s hip that no one else has yet seen, and that he will also rapidly «observe» or deduce all the recondite health problems associated with the woman’s condition. He may also grasp, in light of his medical knowledge, the risks that will ensue in the future if the woman continues to live on her own. If he sees this typical risk clearly, knowing how painful communicating it will be to those present, he may raise his eyes and consider the people around him, gauging the extent to which they too have come to this sad realization. He may now look at the woman not only as a distressed physical body but also as a possible observer of herself, seeking to ascertain how far she is aware of her changed circumstances. He may also glance at the postures of the others in order to gain an idea of their awareness of the problem. By acting in this way, he moves to the second level of observation, which is already an accomplishment for a general practitioner. But he may even shift to the third level if he sees himself as an observer of the observed or, should he be even more acute (for a doctor), he may notice that the neighbour not only sees the facts and problems but also, like him, their connected observation systems.

A doctor is not expected to achieve so much, and the above considerations are only theoretical. If he moves through all these levels as far as the meta-level of «himself observing the observers of the observers looking at the medical fact for which he has been summoned», the doctor would be acting in fact as a social worker. A doctor can be a good doctor, though perhaps not an excellent one, if he knows how to give complete account of the pathology with which he is presented. He is justified if he considers only the objective evidence of clinical relevance. Indeed, he can impose the authority of his specialized knowledge and define the fact or problem as he wishes, even if his definition contradicts the perceptions of the others present. For example, he may decide that the elderly woman does not have any serious medical problem, despite all the blood on her face, whereas the others believe the opposite, and vice versa.

If a social worker observed only the first level of the situation at hand (the emergency of the injured elderly woman) and did not concern himself with the stratified levels of awareness around him, he would quite simply not be doing his job. If the doctor is now joined in the woman’s home by the social worker, the latter could methodically scan all the levels of perception present, even if only to ascertain that those levels are not currently occupied by observers. If instead he realizes that all levels of observation are covered — for instance, if he is lucky enough to be in a situation in which the doctor is able to act as a third-level observer, as envisaged above — then he should be able to see all those levels.5 If the doctor is at the third level, he must see it as such, which means that he must be able to see the two underlying levels and at the same time shift to an even higher one, the fourth.

A social worker is called such because in every problem situation it is his task to ascertain the range of possible perceptions, while standing one degree higher than all those present so that he can survey them. Moreover, a social worker sees him/herself as s/he observes the entire observing system arrayed below him/her. S/he is thus always positioned one level higher than the n observational levels including him/herself, according to the following formula:

social observer = (n observation levels + his/her own) + 1

It follows that social helping work has the formal structure of supervision. This is an important point. For the time being, I merely say that whereas many professions, even highly specialized ones, directly manipulate the facts, social work expresses one competence of observation over other competences of observation, in the same way as a practice teacher or a coach does. The «technique» is the same as that used in advanced vocational training, which obviously includes social work. The sequence is well known. First, the student is confronted with complex tasks (de facto realities). Second, what the student observes-deduces-does is observed by the supervisor, who provides feedback, the purpose of which is to stimulate the student’s reflexivity, so that s/he can resume the task cognitively and emotionally better equipped. The supervisor does not intervene to say what the student must do; even less does the supervisor intervene to do it himself. In this case, it is not a supervisor or a coach — someone who gives instructions or who watches what the student has learned to make him/her learn further — but a manipulator of facts. A social worker in the field acts in the same way by assuming that people in their life-situations are «students» learning to manage themselves and their lives better.


Bhaskar, R. (1978). A realist theory of science. Brighton: Harvester.

Donati, P. (2006). La qualità sociale del welfare. Lavoro sociale, 6 (3), 311-336.

Folgheraiter, F. (2004). Relational Social Work: toward networking and societal practices. London: Jessica Kingsley Publisher.

Giddens, A. (1990). The Consequences of Modernity. Stanford California: Stanford University Press.

Harding, D.E. (1988). The little book of life and death. Penguin: New York.

Husserl, E. (1959). Die Krisis der europäischen Wissenschaften und die transzendentale Phänomenologie. L’Aja: Martinus Nijhoff’s Boekhandel.

1 This general tendency is confirmed, among others, by Donati, who states: «In many cases the reference paradigm [of social workers] is — more or less unconsciously — the medical one — standardized and functional — of the best medical/sanitary practices, according to an old use that considers the medical paradigm as emblematic also for the applied social sciences» (Donati, 2006, p.26, my translation).

2 In the words of Giddens: «Given the greater awareness today that sensory observation is permeated by theoretical categories, philosophical thought has in the main veered quite sharply away from empiricism» (Giddens, 1990, p. 49).

3 Guided by Zen philosophy, Harding says: «Take a close view of X, and it is revealed as one sort of thing; viewed from further off, as quite another sort; from far away, still another. Go on retreating from X, and all traces of it vanish — there’s nothing to see. Come close enough, and again they vanish. Whatever I go up to or away from I lose. [...] I come across nothing that’s absolutely and simply what it is, nothing that isn’t equivocal, relative, phantasmal, indeterminate. [...] It is never something clear-cut and definite: its very nature hangs on how one looks at it: in particular on where one looks at it from, its distance» (Harding, 1988, pp. 16-17).

4 This is an example of paradox identified by Husserl (1959):A man, and in communalization mankind, is subjectivity for the world and at the same time is supposed to be in it in an objective and worldly manner.

5 One may ask how the social worker is able to perceive the woman’s non self-sufficiency immediately upon his arrival. Social workers usually need time to make an assessment, and to do so they must interact with the persons involved: i.e. it is the second-level reality observed (the observing social) which persuades them that the first-level reality (the woman’s non self-sufficiency) exists.