Navigation – Plan du site

AccueilNumérosHS-23The expression of proximity in Fr...

The expression of proximity in French and in English

Claude Vandeloise


La distribution des expressions de proximité en anglais et en français n’est pas déterminée par la distance géométrique mais par l’accessibilité à la cible. Near to ou près de sont choisis si l’accès est facile et far from ou loin de s’il est difficile. Différents facteurs déterminent l’accès à la cible, tout spécialement la vitesse de la cible, celle de l’entité accédant à la cible et la nature du trajet joignant le site à la cible. Cependant, dans des descriptions objectives relevant de la géométrie ou de la physique, la distance évaluée relativement à une échelle dépendant de la taille de la cible et du site établit à elle seule le choix entre near to ou près de et far from et loin de. Le second composant des expressions close to et près de montre clairement que l’anglais et le français adoptent des perspectives différentes concernant l’accès à la cible : to en anglais introduit le but du déplacement alors que de en français indique son origine, l’anglais recourant à la même stratégie dans far from. En conséquence, cette langue oppose to dans close to et from dans far from alors même que le français utilise uniformément de dans près de et loin de. La dernière section de cet article développe le scénario de genèse des prépositions spatiales présenté dans Vandeloise (2010, reproduit dans ce numéro spécial). Elle évalue la position des expressions de proximité dans ce développement.

Haut de page

Texte intégral

1 If contact is considered as an extreme case of proximity, this notion may be considered as a necessary condition shared by all spatial relations. For this reason, one might expect expressions associated to proximity, such as near and close to in English or près de and proche de in French to appear before the other spatial prepositions. This is not so because elementary language is not so much interested in the expression of simple general concepts as it is in complex functional concepts such as containment or support. True, these concepts include contact or proximity among their characteristics. But the words attached to these concepts (in and on in English or dans and sur in French) convey the global notions of containment and support and their marginal variations.

2 The expressions near and près de —and their opposites far from and loin de— evoke the notion of distance between the target and the landmark. In the first section of this article, I will develop a usage rule in terms of ease of access to the target proposed in Vandeloise (1986, 1991). Distance is only one factor contributing to accessibility. Even though the purpose of bearers such as tables is to facilitate access to burdens, one cannot say that *the cup is near the table if the cup is on the table. An explanation to this contrast will be provided in the third section. Proximity is often expressed by complex expressions (près de, near/close to) that can belong to different parts of speech : near and close may be verbs and adjectives and près is often considered as an adverb and proche as an adjective. The second constituents of these expressions (to in close to and near (to) and de in près de) are particularly interesting since they look in opposite directions. Indeed, to expresses the prospective position of the target in English whereas de in French corresponds to from and introduces the origin. As we will see in the second section, this contrast corresponds to different perspectives on the entity that has access to the target. It is often a third entity independent of the landmark. In the third section, I will try to introduce near and près de in the genesis of spatial prepositions I have proposed elsewhere (Vandeloise 2010, reproduced in this special issue). It will appear that, whereas proximity is the most general spatial relationship imaginable, the words devoted to the expression of this notion —and of this notion only— appear very late and are preceded by all the functional spatial prepositions. Whenever the conditions for the use of these prepositions are met, they preclude the use of near or près de.

1. The role of accessibility in the use of near (to), close to, far from and près de, proche de, loin de

  • 1 The use of près without de, as in près la fontaine (example provided by Le Trésor de la Langue Fran (...)

3 There is a parallelism between near (to) and près de on the one hand and close to and proche de on the other hand.1 Indeed, whereas the first pair of expressions can only be used in the spatial and temporal domains, the second pair can also be used in other domains such as color terminology or sentiments :

(1) Le mauve est proche du bleu
(2) Mauve is close to blue
(3) *Le mauve est près du bleu
(4) *Mauve is near blue
(5) Pierre est proche de Peter
(6) Pierre is close to Peter
(7) Pierre est près de Peter
(8) Pierre is near Peter

4Sentences (3) and (4) are unacceptable and sentences (7) and (8) can only receive a spatial interpretation. Together with near (to) and close to, English can also use next to, by and beside to express proximity :

(9) Peter’s house stands next to Paul’s house
(10) Peter stands by the tree
(11) Peter stands beside the house

5The expression next to conveys direct succession in any order. Sentence (9) means that there is no house between Peter’s and Pierre’s house. In a city, this probably implies that the houses are near but this is not necessarily the case in the country. The preposition by is one of the most versatile prepositions in English. It is used mainly when speaking of proximity in the horizontal plane as a less precise alternative for beside, in front of or behind (Lindstromberg 1998). In addition to proximity, the preposition beside (like its French equivalent à côté de) expresses directionality along the lateral axis. According to Carlson and Covey (2005), beside, like near to, is associated with a closer distance than near.

6 As illustrated by sentences (12)-(15), near (to)/far from and près de/loin de cannot be defined in terms of absolute distance, according to which near to and près de would refer to smaller distances than far to and loin de :

(12) Jupiter est près de Saturne
(13) Jupiter is near Saturn
(14) L’électron est loin de son noyau
(15) The electron is far from its nucleus

7Obviously, the distance between Jupiter and Saturn is greater than the distance between the electron and the nucleus. The unacceptability of sentences (18) and (19) reveals the incompatibility of near and près de with the specification of distance. Projective prepositions such as above, in contrast, may be modified by a distance :

(16) The lamp is three feet above the table
(17) La lampe est à un mètre au-dessus de la table
(18) *The lamp is three feet near the table
(19) *La lampe est à un mètre près de la table

8 Rather than defining the proximity expressions by any absolute distance, one may attempt to refer to a contextual usual distance or norm that is different for stars and electrons. A provisory usage rule for near/far from (easily adaptable to près de/loin de) would then be the following :

N’/F’ : a is near/far from b if the distance between target and landmark is lesser/greater than the norm

  • 2 Langacker uses the term trajector for any primary in a relation. Since this article is limited to t (...)

9It is more difficult, however, to specify the qualities of this norm. As we will see, the principal characteristics of the norm relate to the accessibility of the target/landmark to the landmark/target as well as to the speaker or his addressee. Langacker (1987) points out that close to “permits apparently unlimited focal adjustments with respect to scale”. The scale and the norm depend on the trajector2 of the relation. Although the size of the norm is certainly proportional to the dimension of the target, it also depends on different factors :

10(1) The dimension of the landmark : Because of its function as a reference point localizing the target, the landmark is rarely smaller than the target. In this respect, the size of the target often determines the minimum norm of the relation. But the norm may also increase in proportion to the landmark. Thus, the distance between Jupiter and Saturn is greater in sentence (20) than in sentence (12), because the Milky Way is larger than Saturn :

(20) Jupiter is near the Milky Way

11(2) The speed of the target : If the target is moving toward the landmark, the proximity expressions evaluate the ease or the difficulty of the route. In this case, the norm may increase with the speed of the target. In sentences (21) and (22), illustrated by figure 1, the distance between the target and the wood seems larger for the rabbit than for the tortoise :

(21) The tortoise is far from the wood
(22) The rabbit is far from the wood

12If, on the other hand, the target is moving away from the landmark (figure 2), the norm will increase as speed decreases. Therefore, in the hunting expeditions described in sentences (23) and (24), the distance between the hunter and the tortoise may be greater than the distance between the hunter and the rabbit :

(23) The tortoise is far from the hunter
(24) The rabbit is far from the hunter

13The hunter must be further from the tortoise than from the rabbit in order to make the hunter’s access to the tortoise more difficult. In other words, the normal distance increases/decreases if the target’s speed makes its meeting with the landmark easier/more difficult.

14(3) The speed of the landmark : Landmarks moving relative to the target are uncommon in spatial relations. However, in sentences (25) and (26), illustrated by figure 3, the normal distance increases with the speed of the landmark :

(25) The wounded man is far from the helicopter
(26) The wounded man is far from the Saint Bernard

15In the hunting scenes described in sentences (27) and (28), illustrated by figure 4, the landmark is moving away from the target. The distance considered the norm between the two animals varies depending on the speed of the landmark : this distance is smaller between the fox and the rabbit than between the fox and the hen :

(27) The fox is near the rabbit
(28) The fox is near the hen

16Hence, the norm depends on the speed of the target relative to the landmark (figure 2) as well as on the speed of the landmark relative to the target (figure 4). If speed makes an encounter between these two entities easier/more difficult, the normal distance increases/diminishes. Notice that in figure 2 and 4, the purpose of sentences (23), (24), (27) and (28) is not so much to localize a target relative to a landmark as to describe the action of an agent that tries to reach a patient.

17(4) The size of the speaker : A toy in the middle of a table may seem far from the edge to a child, but near to his father. Once again, the ability to contact the target plays a role in determining the normal distance.
(5) The speed of the speaker : A bar may seem far from the church if the speaker is on foot, but near if he is driving. The choice of these expressions is determined by ease of access rather than by distance.
(6) The size and speed of the addressee : If the addressee is involved in situating the target, the speaker can take his addressee’s possible access to the target into account. Thus, a house will be considered near a village if the addressee is driving but far if he is walking.
(7) Facility of access : The normal distance increases/diminishes if the path to the meeting place of target and landmark is easy/difficult. For example, two houses on a hill at equal distance from the speaker may be judged near or far according to whether the speaker must walk up or walk down.
(8) Types of access : The norm is essentially determined by physical access. However, access to perception may also play a role. Therefore, a mountain may be near if we admire its beauty from a window but very far if we have to climb it. Auditory access and physical access may also be in opposition : wolves may seem near when we hear them howling in the woods, but far enough not to threaten our lives. Finally, a boat may be far to the naked eye, but near through binoculars.

18 All these factors clearly show that the target’s/landmark’s access to the landmark/target is determinant in the use of near and far from, as well as their French equivalents près de and loin de. The first usage rule N’/F’, as a function of the normal distance, hides the role of accessibility behind the role of the norm. However, accessibility alone can provide an exhaustive description of the distribution of proximity expressions. Therefore, instead of usage rule N’/F’, I propose usage rule N/F :

  • 3 The terms in parenthesis are not included in the usage rules proposed in Vandeloise (1986, 1991) fo (...)

N/F : a is near/far from b if the target/landmark is easily/not easily accessible to the landmark/target (or to a third implicit entity located at the position of the landmark)3

19In logical diachrony (Vandeloise 1986, 1991), this rule may be considered as an impetus, a first meaning from which the extensions of the preposition develop. All the factors facilitating or complicating the encounter between target and landmark play a role in the impetus of near/far from. Obviously, distance is one of these factors : everything being equal, a target is nearer the landmark if the distance diminishes and further if it increases. In most objectively descriptive situations such as those described by sentences (12)-(15), distance takes an important lead and is the only factor determining the choice of proximity expressions. Therefore, in order to account for these cases, I will add an extension N1/F1 to the impetus N/F of near/far from. Any reference to a norm of distance must be avoided in this extension since it would reintroduce all those other factors determining access. The sizes of the target and the landmark in descriptively objective situations establish alone a scale according to which the distance between target and landmark may be evaluated. The extension N1/F1 canbe formulated as follows :

N1/F1 : a is near/far from b if the distance between the target and the landmark is small/large according to the scale determined by their dimension

20 The impulsion N/F of near/far from and its extension N1/F1 correspond to the analysis proposed in Vandeloise (1986, 1991) for the French equivalents of these expressions, près de/loin de. This analysis calls for elaboration. First, access can occur between the target and the landmark only if the target is animate : can a spoon near a cup access the cup ? The second constituent of the proximity expressions near (to), close to and près de will provide important information on the perspective adopted by the third entity. The contrast between to, introducing the prospective position of the target in English, and de, introducing origin in French, is especially interesting. The second problem for the analysis proposed in this section is that easy access to the target from the landmark does not always allow for the use of proximity expressions. Indeed, sentences (29) and (31) describe situations in which the target can easily be accessed from the landmark but sentences (30) and (32) are unacceptable if the apple is contained in the basket or borne by the table :

(29) The apple is in the basket
(30) *The apple is near the basket
(31) The apple is on the table
(32) *The apple is near the table

  • 4 In this way, proximity expressions behave like the verb to touch (Vandeloise 1996) relative to verb (...)

21In other words, near cannot be used when a functional relation occurs between the target and the landmark as a consequence of their proximity.4 I will deal with this question in section 3.

2. Perspectives in the expressions of proximity

22 In this section, I will establish that proximity expressions in English and French present different perspectives on the situations they convey. These perspectives are introduced by an entity that may be different from the target or the landmark. Only functional spatial prepositions such as in, on, at and under exclusively involve the target and the landmark : an apple is in, on, or under a basket whether or not there is an observer to notice its position. Projective prepositions such as above, in front and on the left could not be interpreted without the existence of a third entity : the vertical, frontal and lateral axes. Among these axes, the vertical axis is the most objective since it does not depend on the orientation of the speaker. Frontal and lateral axes are determined by the orientation of the speaker or by a landmark with an intrinsic orientation, such as a chair or a cupboard. If such is not the case, the speaker or another intrinsically oriented entity must provide the landmark with a contextual orientation. Thus, in addition to the frontal and lateral axes, a further entity is indirectly introduced in the relation. Besides its projective uses, behind may mean that the landmark hides the target from a third entity —very often the speaker. A third entity is therefore always involved in this relation. We will see that this is also true of the complex proximity expressions, close to and près de. In contrast to between and among, which involve more than two explicitly expressed entities, the third entity of behind and the proximity expressions is always implicit.

  • 5 May 2, 2007, I found 501,000,000 occurrences of near without to for 2,170.000 occurrences of near t (...)

23 The most frequently used proximity expression in English is near. It behaves like a simple preposition with the usual target/landmark asymmetry : the target is normally smaller and more mobile than the landmark. As we saw in section 1, however, the sizes and the speeds of the speaker and the addressee may play a role in the choice between near and far from. The complex expression near to is much less often used than near (less than 1 % on Google5). It appears mostly with a notional meaning (near to God) and before verbs and gerundives (near to breaking). Even though close to is also less frequent than near and often occurs with notional meanings, this complex proximity expression is the best place, with far from, to observe the perspective imposed by proximity expressions in English. In this section, in keeping with the “localist hypothesis”, I will assume that to in close to introduces the prospective final position of the target in a movement whereas from in far from (and de in près de and loin de) introduce its origin. Then, I will see which movements are suggested by the use of proximity expressions in French and English.

24 The relation of to with an abstract movement is easy to justify for the temporal and notional uses of close to in sentences (33) and (35) :

(33) Christmas is close to New Year’s
(34) ?New Year’s is close to Christmas
(35) Peter is close to his family

25Indeed, in time, the present may be conceptualized as moving toward the future. In sentence (33), to introduces New Year, the term of the abstract movement relative to which Christmas is located. Sentence (34), in which the target takes place before the landmark, seems awkward. For the notional use in sentence (35), the family may be considered as the end toward which Peter aims his affection.

26 The justification of to in the spatial uses of close to is more complex. I will consider in turn the uses of close to involving at least one moving entity and those involving two still entities. Sentence (36) describes a moving target and a moving landmark whereas the target only is moving in sentence (37). Sentence (38), in which the landmark only is moving, seems anomalous without appropriate circumstances :

(36) The fox is close to the hen
(37) The fox is close to the henhouse
(38) ?The henhouse is close to the fox
(39) The fox is wounded but, fortunately, the henhouse is close to the hungry animal

27In sentence (36), the preposition to may be justified relative to the trajectory of the fox and the hen : the fox is going to the hen and its position on the trajectory allows an easy access to the hen. In sentence (37), the trajectory justifying the use of to is determined by the fox only. Even though the running fox also traces a trajectory in sentence (38) as well, this sentence seems anomalous because of the usual target/landmark asymmetry. In sentence (39), however, the circumstances reinforce the role of the henhouse as a goal whose proximity is vital for the fox and facilitates the use of close to. The use of close to with at least one moving entity is illustrated by figure 5 :

28 No explicit trajectory can be found in sentences (40) and (41) :

(40) The spoon is close to the cup
(41) The tree is close to the church

29One might say that the position of the spoon is the result of a preliminary move towards the cup but this option does not stand for intrinsically immobile entities like the tree and the church. Another possibility might be to evoke the displacement of a third entity moving between the target and the landmark. But, since the landmark is the point of reference localizing the target, a movement from the landmark to the target is much more likely than the reverse. Thus, the landmark would be the point of departure of the displacement and from should be used instead of to, as it is the case in French for près de. Another difficulty for an explanation of close to by prospective trajectories appears in figure 6 :

30Both speakers in figure 6 are likely to use sentence (41) rather than sentence (42) because of the target/landmark asymmetry :

(42) The church is close to the tree

31But the trajectory of Sp1, who meets the church before the tree, can justify the use of to in sentence (41) whereas the trajectory of Sp2 is more compatible with sentence (42). Hence, the use of to when close to relates immobile entities cannot be justified by a prospective trajectory. Furthermore, it contradicts the target/landmark asymmetry according to which the normal way to take possession of the target is to go first to the landmark and then from the landmark to the target. Therefore, one may surmise that the presence of to in close to is founded on the temporal and notional uses of close to and on the spatial usages of this expression in which at least one of the related entities is mobile. This expression is then globally extended to the occurrences of close to for which both entities in relation are immobile. Whereas to favors mobile entities, we will see that from in far from and de in près de and loin de are more compatible with immobile entities.

32 In contrast to French, which uses the same preposition de in près de and loin de, English uses to in close to, but from in far from. If this expression were interpreted according to the same perspective as close to, figure 7 would represent the relation between the two expressions :

33But, in this case, the landmark would be interpreted as the goal of the target and *far to would fit this conceptualization better than far from. Of course, the landmark of far from is often the goal at which the target is aiming. But this is less obvious than with close to as illustrated by sentences (43) and (44) :

(43) The car is too close to the truck
(44) The car is too far from the truck

34In sentence (43), too before close to reinforces the possibility of an encounter, probably undesirable if not harmful, with the truck. Before far from, in contrast, too means that the car is unlikely to reach the truck and the goal interpretation of the landmark is cancelled. Therefore, rather than as a goal, the landmark of far from must be interpreted as a point of reference from which the possibilities of encounter between the target and the landmark are evaluated, as in figure 8 :

35Of course, if the spoon is far from the cup, the cup itself does not evaluate its accessibility by the spoon. This interpretation necessitates a mental transfer of the speaker to the position of the landmark. The representations of the perspectives in sentences (45) and (46) are rather complex :

(45) Peter is close to the tree but Pierre is far from it
(46) Peter is close to the tree but far from the church
(47) Peter is near the tree but Pierre is far from it
(48) Peter is near the tree but far from the church

36These sentences are represented by figures 9 and 10 :

37According to the ‘localist’ hypothesis adopted here, two points of reference are necessary to conceptualize sentences (45) and (46) : the end of the virtual movement of Peter and the origin of the virtual movement of Pierre in the former sentence ; and the end and the origin of the virtual movement of Peter in the later. Sentences (47) and (48) do not require the same complexity, since near does not necessarily introduce the end of the movement in the picture. This may explain why near is used more frequently for spatial proximity and close to for temporal and notional uses in which the role of the preposition to is more transparent. This preposition is also more transparent when close to relates spatial moving entities than with still entities. If, in its spatial uses, close to were more frequent with moving entities than with immobile entities, this would confirm the above analysis.

  • 6 The role of de in x est à gauche de y is less clear. It might very well introduce the point of refe (...)
  • 7 Gunnarson (1986) analyses this sequence as [[près]Adv[de la table]PP]AdvP, in which Adv and AdvP me (...)

38 All proximity expressions in French are complex and, in contrast to English, they share the same second component de in près de, proche de and loin de. This component does not have the same meaning as de in the projective expressions au-dessus de (‘above’) and à la gauche de, in which de specifies the noun phrases le dessus and la gauche.6 Whereas the latter might be translated by ‘at (the) left of’, de should be translated by from in the proximity expressions près de, proche de and loin de : ‘near from’, ‘close from’ and ‘far from’. The structure of French proximity expressions is also different from the construction of projective expressions. Since au corresponds to the fusion of à + le, the structure of au-dessus de la table is [[à]Prep[[le dessus]NP[de]Prep[la table]NP]NP]PP, in which Prep means “preposition”, NP “noun phrase” and PP “prepositional phrase”. This means that de contributes to creating a complex noun phrase after the preposition à. In près de la table, in contrast, the structure is [[près de]PL[la table]NP]PP,7 in which PL means “prepositional locution”.

39 The discrepancies between French and English reveal that proximity expressions are in a zone of turbulence in which different and competing motivations can trigger different choices such as to in close to instead of de (‘from’) in près de. It should be pointed out that French has fluctuated in its choices, since près à (like approcher à instead of approcher de) is attested in Old French (Vising, quoted in Clédat 1927). The use of to in close to may be an adequate choice for the temporal and notional uses of this expression, as well as for its spatial uses relating moving entities. However, its interpretation is more difficult for spatial uses relating static entities and the contrast between close to and far from creates problems in the interpretation of sentences in which these two expressions appear together (see figures 9 and 10). In French, on the contrary, the choice of de allows for a direct interpretation of the static spatial usages of proximity expressions but may be problematic for the temporal and notional uses of proche de as well as for the spatial uses relating moving entities.

40 In sentences (49) and (50), the landmark introduced by de may be interpreted as a point of departure or a vantage point from which the accessibility to the target is evaluated :

(49) Le crayon est près de la lampe
(50) La cuiller est loin de la tasse

41Sentence (49) might be paraphrased by le crayon est près de la lampe à partir de la lampe. The landmarks in sentences (49) and (50) are static and inanimate and cannot have access to the target. The interpretation of these sentences involves a third entity, very often the speaker, who makes a mental transfer to the position of the landmark to evaluate the ease of access to the target. This may be even more obvious in sentence (51) in which the most likely trajectory of the speaker to access Pontoise is described in figure 11 :

42 (51) Pontoise est près de Paris

43Of course, a driver provided with the information in sentence (51) comes across a signpost leading to Pontoise on his way to Paris and go directly to his destination (figure 12).

44Nevertheless, figure 11 will fit his trajectory in a majority of cases.

45 When the entities related by the proximity expressions are moving, as in sentences (52) and (53), French does not focus on motion as English does, but evaluates the ease of access of the target to the landmark from the landmark’s point of view :

(52) Le renard est près de la poule
(53) Le lapin est loin de la tortue

46This perspective may seem less natural than the perspective adopted by English, but English proceeds in the same way with far from. Furthermore, the use of de in both près de and loin de allows for a common representation of these expressions in figure 13 :

47The expression of proximity can be considered as a zone of turbulence for which competing motivations introduce different conceptualizations. This means that there is no single way of conceptualizing the situation for speech and that different languages can choose different alternatives. For every advantage offered by any one choice, there is a price to pay and disadvantages to accept.

48 The use of de (‘from’) in the temporal and notional uses of proche de in sentences (54) and (55) is also less motivated than the use of to in close to.

(54) Noël est proche du Nouvel An
(55) Pierre est proche de sa famille

49Any attempt to justify the choice of de in proche de by the conceptualization of the situations conveyed by sentences (54) and (55) is doomed to failure. The only option is to call for help upon the system of proximity expressions in which de is first motivated for their localization uses, and then to extend it by analogy to the spatial uses relating moving entities and to the temporal and notional uses of proche de.

3. The position of proximity expressions in the genesis of spatial prepositions

50 Conceptually, proximity (with coincidence as an extreme case) is the most general spatial relation, which is included in all the other spatial relations, even if it is to be denied as in far from. Therefore one might expect this concept to be among the first to be given linguistic expression. In fact, we will see that its linguistic expression appears late in the genesis of prepositions that is suggested here. This is confirmed by the fact that prepositions expressing proximity or coincidence involve an implicit third element in contrast to functional prepositions such as in and on, which only engage two entities, as well as by their complex morphology involving more than one component. One of these components often belongs to another class than prepositions, such as the verbs or adjectives close and near in close to or near (to) ; the adverb près in près de ; and the adjective proche in proche de. This discrepancy between the overall abstract simplicity of the notion of proximity and the complexity of its expression will come as a surprise only to those who assimilate languages to formal systems evolving from simple to complex concepts. On the contrary, it is my view that the first situations singled out for linguistic expression are recurrent functional relations essential for our survival in the world, such as containment and support. As far as their purpose is concerned, these relations are very straightforward. But the means by which this purpose is fulfilled are multiple. All the characteristics required by the fulfillment of the relation appear in prototypical cases, whereas marginal situations meet only some of the characteristics. This explains why, even though proximity is a characteristic of every spatial relationship, it takes time before a language feels the need to express proximity per se, independently from the other characteristics that contribute to the structuring of the spatial linguistic expressions.

  • 8 Adpositions include prepositions and postpositions.
  • 9 See figure 18 p.512 in Levinson and Meira (2003) for a complete chart.

51 Levinson and Meira (2003) were the first to propose a model for the development of spatial terms in languages. This development is based on the fact that many languages have only one term to designate spatial relationships. Levinson and Meira claim that further spatial terms appear in a fixed order in the development of languages. Like Berlin and Kay (1969), they use a sample of genetically unrelated languages to justify their analysis. Informants in each language were asked to ascribe an adposition8 in their language to a booklet of 71 line-drawings known as the “topological relations picture series”. As in the case of the attribution of basic color terms, the choices of the pictures tended to cluster and were not randomly distributed as they would be if there were no cross-linguistic generalizations. On the basis of their data, Levinson and Meira propose an implicational scale according to which spatial terms develop. I only mention the terms that are useful to understand the position of NEAR in their analysis :9

52AT1 is a unique spatial notion that covers all the spatial relations and corresponds to the adposition ta in a language like Tzeltal or di in Indonesian (Feist 2004). AT2, AT3 and AT4 are more and more specific notions : AT2 covers all spatial relationships except those conveyed by IN and AT4 is a residue that excludes all the preceding notions, including NEAR. In chart 1, this notion is last to receive a linguistic expression.

53 In Vandeloise (2010, reproduced in this special issue), I discuss Levinson and Meira’s analysis before arguing for an alternative solution. This alternative does not integrate proximity expressions. I will first present a slightly modified version of this analysis. Then I will try to incorporate proximity expressions in the analysis. In my alternative, I claim that the genesis of spatial prepositions is grounded in a process of lexical formation by division (MacLaury 1991). In this process, a new appellation is attributed to a group of prominent members of an already existing lexical category. First, the new appellation coexists with the original appellation. After a while, the new appellation supersedes the old one for the prominent members of the category and the old appellation is restricted to the other members. When the division is complete, a new lexical category is created. Originally, one may surmise that a single word —let us call it X— was used for all spatial English and French relations, as is still the case for ta in Tzeltal (Brown and Levinson 1992) or di in Indonesian (Feist 2004). Among the spatial relationships covered by X, speakers identify a group of prominent relations, for example containment. Let us call IN the word attached to this category. At first, IN is used jointly with X by the younger or more creative speakers each time a further clue is necessary to indicate containment and dissipate the ambiguity with the other relations. Later, the use of IN instead of X becomes mandatory. At this point, the general lexical category X is split into two categories : X1, including all the spatial relationships except those conveyed by IN and IN. Figure 14 represents this development. At stage 1, X can always be used instead of its hyponym IN. At stage 2, IN becomes mandatory and the extension of X1 covers all spatial relationships except those of containment.

54If this development is true, IN at stage 1 may be considered as an elaboration of the general preposition X. At stage 2, however, X1 and IN are independent categories. In fact, the possibility of using IN makes the use of X1 impossible. This mutual exclusion will be considered as a criterion for lexical formation by division :

Criterion for lexical formation by division : A name attributed to a category created by lexical division can no longer be applied to the general category

55This criterion will serve as a guideline for the incorporation of proximity expressions in the genesis of spatial terms. It distinguishes lexical formation by division from lexical formation by union, instrumental in the formation of super-categories like animals or tools. In this case, the formation of the word animal does not preclude the use of basic words like dog or cat to designate the animal.

56 Extending the process of lexical extension by division, one may propose the following development for functional spatial relationships in French :

57The prepositions dans and sur are functional prepositions that convey containment and support, that is to say dynamic relations in which the landmark controls the target in more than one direction or along the vertical axis respectively. I have introduced sous between sur and à because the landmark of this preposition does not exert a constraint over the target when these entities are not in contact. Thus, there is no necessary dynamic connection between the entities related by sous. However, in any circumstances, the landmark of sous prevents accessibility to the perception of the target. Therefore, sous is more dynamic than à but less functional than dans and sur. The French preposition à in the chart has exclusively its localization meaning. This means that the landmark helps to localize a target —often located outside the visual field— without dynamic constraint. The landmark often designates a geographical area with a part of which the target coincides. As indicated by the arrow in chart 2, the preposition à can also be used when a target participates in a routine associated to the landmark (Vandeloise 1988), as in Pierre est à l’école. This use that does not require that Pierre be at school at the time of speaking should be considered as a later extension that remains internal to the preposition, unconnected to its introduction in the genesis of spatial relationships. Such further extensions are frequent in the development of prepositions from their original meaning in the chart. From a synchronic perspective, they blur the development of spatial prepositions by lexical division. This is especially true of English. In this language, there is an extension of in from the expression of containment to the expression of localization, originally expressed by at, coming from Old English œt. Therefore, occurrences of in are in contrast with X4, together with at (Vandeloise 2008, reproduced in this special issue).

58 X1, X2 and X3 are residues created by the separation of dans, sur, sous and à from X. The development of the last residue, X4 will be essential for the development of proximity expressions. At this stage, X4 applies to all spatial relationships except dynamic relationships —conveyed by dans and sur—, relationships involving concealment —conveyed by sous—, and relationships implying partial coincidence between the target and the landmark —conveyed by à. In positive terms, this means that X4 conveys all the spatial relationships that do not imply functional connections or coincidence between the target and the landmark. All these spatial relationships may be conveyed by the proximity expressions near and far from and près de or loin de. Therefore, it would be very tempting to associate X4 directly with proximity expressions. But projective prepositions in the vertical, frontal and lateral axes can also be associated with these relationships with a more specific meaning. Thus, the order of appearance in the genesis of spatial terms of proximity and projective expressions remains to be determined. The criterion of exclusion is crucial for this choice.

59 In lexical formation by division, the use of a preposition for a situation precludes the use of all the prepositions lower in the chart : one cannot say that a stone in a box is on the box or at the box. Notice that one cannot say that it is in front of the box or near the box either. This suggests that projective and proximity expressions appear after functional prepositions in the genesis of spatial terms. As far as the possibility of using projective expressions for an identical situation is concerned, there is a difference between the expressions in the vertical axis and in the horizontal axes. Whereas a situation described by sentence (56) is difficult to describe by means of sentence (57), sentences (58) and (59) as well as sentences (60) and (61) apply without difficulty to an identical situation if the target is easily accessible from the landmark :

(56) The lamp is above the table
(57) ?The lamp is near the table
(58) The tree is in front of the house
(59) The tree is near the house
(60) The tree is on the left of the house
(61) The tree is near the house

60However, the substitution of above by near is not as anomalous as the substitution of in by on or of on by at would be to describe a prototypical situation of containment or support. Therefore, one may wonder if sentences (56) and (57) provide a sufficient argument for lexical formation by division. Furthermore, in contrast to functional prepositions dans, sur, sous and à, the expressions au-dessus de/en dessous de come in pairs like the other projective and proximity expressions. Tentatively, however, I will introduce these expressions in chart 3 :

61In chart 3, X5 conveys all the horizontal spatial relationships that do not imply functional connections or coincidence between the target and the landmark. These relationships may be conveyed by in front of/behind, on the left/on the right and next to/far from in English and devant/derrière, à gauche/à droite and près de/loin de in French. Sentence (64), including a proximity expression, can be used to describe the situations expressed by projective expressions in sentences (62) and (63), provided ease of accessibility to the rabbit is granted :

(62) The rabbit is in front of the tree
(63) The rabbit is on the left of the tree
(64) The rabbit is near the tree

62Therefore, the criterion for lexical formation by division of the residue X5 is not met. There cannot be a relation of formation by lexical division between in front of and on the left either since these expressions can combine to make the description of a single situation more specific :

(65) The rabbit is in front of the tree, on the left

  • 10 The scope of x is beside y (or x est à côté de y in French) is more restricted than the scope of x (...)

63I will propose three hypotheses for the development of proximity and horizontal projective expressions from the residue X5. First, proximity expressions are more general than the projective expressions and next to/far from or près de/loin de alone can describe all the situations belonging to X5.10 This suggests the following development :

64According to chart 4, proximity expressions would appear before the horizontal projective expressions that specify this relation. The reverse order is proposed in the second hypothesis :

65Now, the situations in X5 first receive a specific name along the frontal and the lateral axis. Then, proximity expressions appear as supercategories, by lexical formation by union. According to the third and last hypothesis, proximity expressions appear independently at a late stage in the development of spatial prepositions. This hypothesis is made plausible by the fact that proximity expressions are often complex and constituted by components belonging to different grammatical categories.

  • 11 This term subsists in topographic expressions.

66 Diachronic data might provide precious clues to choose between these hypotheses. As they stand now, they provide no clear evidence. In French, devant (de + avant) is first attested ca 1050 (TLF), at the same time as près de (from the Latin adverb presse, derived from the past participle pressum of the verb premere, ‘hold tight’). As for derrière (from derere), it is first attested a little later, ca 1100 (TLF). Proximity in Old French was also expressed by three groups of expressions that have now disappeared (coste, encoste, decoste ; jouste, dejouste, pardejouste ; and lez11, delez, par delez) as well as by ras de (from the Latin participle rasus). Fagard (2006) provides detailed data on the evolution of coste, jouste, lez and their derived expressions but says nothing about the development of près. Thus, our present knowledge of diachronic data cannot help to select the correct hypothesis. Hopefully, the proposed hypotheses can provide clues to historical linguists who often have to work with fragmentary and incomplete data.

4. Conclusion

67 The distribution of proximity expressions in English and in French is not determined by geometric distance but by accessibility to the target. Near or près de are chosen if access is easy and far from or loin de if it is difficult. The entity accessing the target may be the landmark when it is animate. Otherwise, a third entity —often the speaker— transfers itself to the landmark’s position and evaluates the accessibility from this point of view. Various factors determine the access to the target, notably the speed of motion of the target, the speed of the entity accessing the target and the nature of the route joining the landmark to the target. Of course, if all these conditions are similar, ease of access is determined exclusively by distance. In objective descriptions, such as those of geography or physics, distance evaluated relative to a scale determined by the size of the target and the landmark establishes alone the choice between near and far from. Ease of access determines the impetus N/F of near/far from from which their total distribution emerges. An extension N1/F1 accounts for the most objective uses of these expressions.

68 The second components of the expressions close to and près de show clearly that English and French take different perspectives on the access to the target : to in English introduces the goal of a movement whereas de ‘from’ in French indicates its origin, with English adopting the same strategy in far from. As a result, this language opposes to in close to and from in far from while French consistently uses de in près de and loin de. These discrepancies show that the relation of accessibility can be conceptualized in different ways. Accessibility can concern moving entities, as in the fox is close to its prey, or static entities, as in the spoon is close to the cup. With moving entities, to is consistent since the prey is the goal of the fox. In this case, the prey is a patient rather than a landmark helping to localize the fox. Things are different with the spoon and the cup. Here, the cup situates the spoon and access goes from the landmark’s position to the target, motivating the use of de in près de. In this way, the perspective adopted by English applies directly when proximity expressions are used with moving entities as well as in the temporal domain (Tuesday is close to Wednesday) and in the notional domain (Peter is close to Pierre). In the case of immobile entities such as the spoon and the cup, to can only be justified as an extension of the preceding usages. The reverse is true of de in French, which applies directly to immobile entities but must be extended for mobile entities and in the temporal and notional domains. Whenever there are competing motivations for the choice of a word, each option presents advantages and disadvantages. It should be noted that the discrepancy between the second components of close to and far from in English also forces the choice of different perspectives in a sentence like Pierre is close to the tree but Peter is far from it.

  • 12 The fact that most proximity expressions are complex cannot be considered as an argument in favor o (...)

69 The last section of the article elaborates on a genesis of spatial prepositions proposed in Vandeloise (2010, reproduced in this special issue). It evaluates the position of proximity expressions in this development. The genesis begins with a single preposition, contrasting all the spatial relations with the other domains in the lexicon. The only common characteristic shared by these relations is proximity between the related terms. Therefore, one might expect proximity expressions to appear at the first stage of the evolution. This is not so, be it only because expressions of proximity, such as near and près de, do not apply when there is contact, which, notionally, is a paradigmatic case of proximity. The evolution from the general spatial preposition proceeds first by lexical formation by division, according to which the most salient spatial relations detach themselves from the bulk of spatial relationships and are associated to new prepositions. These new prepositions first coexist with the general term and then exclude it. The first spatial relations to emerge are not simple general notions such as proximity but functional relations such as containment and support, attached in English to in and on and in French to dans and sur. The function of these relationships and their prominent significance in our everyday life guarantee their coherence. However, they manifest themselves through a rich and complex set of characteristics. All of them are met in the prototypical cases to which the prepositions are first attached. Later, these prepositions develop and can be used when only some of the characteristics are met. The priority of complex functional notions comes as a surprise only if one expects languages to proceed from the descriptively simple notions to the more complex ones. In fact, languages are devised to help our adjustment to the world and to society. Functional relationships fit this purpose better than general abstract notions. The late development of proximity expressions is confirmed by the fact that, beside the target and the landmark, they often involve a third implicit entity necessary to evaluate the accessibility of the target. Also, proximity expressions are complex prepositions and some of their components such as close (a verb or an adjective) in close to or proche (an adjective) in proche de belong to other parts of discourse than prepositions. Even the simple preposition near may be a verb or an adjective.12

70 Various alternatives have been left open in the genesis of spatial prepositions. A choice between them requires further research. First, four residues X1, X2, X3 and X4 are created from the most general relation X by detachment of containment, support, obstruction to perception and localization. Whether or not the detachment of the projective prepositions in the vertical axis, above and below creates a fifth residue X5 awaits confirmation. There are arguments to show that projective prepositions in the horizontal plane (such as in front and on the left), as well as proximity expressions, do not develop by detachment from X5. Indeed, the criterion for lexical formation by division is not met, since the possibility of using one of these expressions to describe a situation does not preclude the use of the other expressions as it does with lexical formation by division. The exact path according to which the development occurs, however, is left open. Three hypotheses are proposed : (1) proximity expressions are directly attached to X5 ; (2) X5 is divided between expressions in the frontal direction (such as in front) and the lateral direction (such as on the left) and proximity expressions emerge as supercategories by lexical formation by union and (3) proximity expressions are independent from the system of spatial prepositions, an assumption made plausible by the fact that many of their constituents are not prepositions. In the absence of historical data to confirm one of these hypotheses, one may hope that they can help to circumscribe and facilitate the diachronic research that is condemned to rely on fragmentary data.

71 In view of these observations, the usage rule N/F for near/far from needs to be refined :

N/F : a is near/far from b if the target/landmark is easily/not easily accessible to the landmark/target (or to a third implicit entity located at the position of the landmark)

72Indeed, as this rule is formulated, near could be substituted to on in the apple is on the table since the landmark makes the access to the target easy. Such is obviously not the case. Two solutions may be offered to this problem. According to the former, the genesis of spatial prepositions in Chart 3 imposes constraints on the choice of prepositions : even though a situation meets all the requirements imposed by the usage rule of a spatial expression, this expression cannot be applied if the situation satisfies the requirements of the usage rule of an expression higher in the hierarchy. The second option would be to modify rule N/F as follows :

a is near/far from b if the only connection between a and b is that the target/landmark is easily/not easily accessible to the landmark/target (or to a third implicit entity located at the position of the landmark)

73Obviously, this formulation is not very palatable. Furthermore, it would wrongly prevent the use of proximity expressions for situations in which horizontal projective expressions can be used. Indeed, more conditions must be met in order to say that a dog is in front of a tree than in order to say that the dog is near the tree. However, the latter sentence is perfectly acceptable to describe this situation. Hence, recourse to the genesis of spatial prepositions avoids a very awkward and cumbersome formulation of the usage rule for proximity expressions.

Haut de page


Berlin, B. & Key, P. (1969). Basic color terms : their universality and evolution. Berkeley & Los Angeles : The University of California Press.

Brown, P. & Levinson, S. (1992). ‘Left’ and ‘right’ in Tenejapa : investigating a linguistic and conceptual gap. Zeitschrift für Phonetic, Sprachwissenschaft und Kommunikationforschung, 45(6) : 590-611.

Carlson, L. & Covey, E. (2005). How far is near ? Language and Cognitive Processes, 20 : 617-631.

Clédat, C. (1927). Près de et approcher de. Revue de Philologie Française, t. 39 : 140-141.

Fagard, B. (2006). Evolution sémantique des prépositions dans les langues romanes : illustrations ou contre-exemples de la primauté du spatial ? Thèse de Doctorat : Université Paris VII.

Feist, M.I. (2004). Talking about space : a cross-linguistic perspective. In K.D. Forbus, D. Gentner & T. Regier (Eds.), Proceedings of the Twenty-Sixth Annual Meeting of the Cognitive Science Society. Mahwah, NJ : Lawrence Erlbaum, 375-380.

Feist, M.I. (2010). Inside in and on : typological and psycholinguistic perspectives. In V. Evans and P. Chilton (Eds.), Language, cognition and space : the state of the art and new directions. London : Equinox, 95-114.

Gunnarson, K.A. (1986). Loin de X, près de X et parallèlement à X : syntagmes prépositionnels, adjectivaux ou adverbiaux. Le Français Moderne, 54(1/2) : 1-23.

Langacker, R.W. (1987). Foundations of Cognitive Grammar I : theoretical prerequisites. Stanford, CA : Stanford University Press.

Levinson, S. & Meira, S. (2003). Natural concepts in the spatial topological domain –adpositional meanings. Language 79(3), 485-516.

Lindstromberg, S. (1998). English prepositions explained. Amsterdam : John Benjamins.

MacLaury, R. (1991). Social and cognitive motivations of change : measuring variability in color semantics. Language, 67(1) : 34-62.

Trésor de la Langue Française, computerized version (TLFi, 2004). Nancy : ATILF/CNRS.

Vandeloise, C. (1986). L’espace en français : sémantique des prépositions spatiales. Paris : Editions du Seuil.

Vandeloise, C. (1988). Les usages spatiaux statiques de la préposition à. Cahiers de Lexicologie, 53 : 119-148.

Vandeloise, C. (1991). Spatial prepositions : a case study in French. Chicago, IL : The University of Chicago Press.

Vandeloise, C. (1996). Touching : a minimal transmission of energy. In E. Casad (Ed.), Linguistics in the redwoods. The expansion of a new paradigm in linguistics. Berlin : Mouton De Gruyter, 541-567.

Vandeloise, C. (2008). Three basic prepositions in French and in English : a comparison. Carnets de Grammaire, 19. Toulouse : CLLE-ERSS report.

Vandeloise, C. (2010). Genesis of spatial terms. In V. Evans and P. Chilton (Eds.), Language, cognition and space : the state of the art and new directions. London : Equinox, 171-192.

Haut de page


1 The use of près without de, as in près la fontaine (example provided by Le Trésor de la Langue Française) is obsolete.

2 Langacker uses the term trajector for any primary in a relation. Since this article is limited to the spatial domain, I find more convenient to use the term target.

3 The terms in parenthesis are not included in the usage rules proposed in Vandeloise (1986, 1991) for the French correspondents of near/far from.

4 In this way, proximity expressions behave like the verb to touch (Vandeloise 1996) relative to verb of action. Indeed, to touch can only be used if the exchange of energy involved by the action is minimal.

5 May 2, 2007, I found 501,000,000 occurrences of near without to for 2,170.000 occurrences of near to.

6 The role of de in x est à gauche de y is less clear. It might very well introduce the point of reference y from which the position of x is evaluated but if that is the case, de should be seen as having different meanings in à la gauche de and in à gauche de.

7 Gunnarson (1986) analyses this sequence as [[près]Adv[de la table]PP]AdvP, in which Adv and AdvP mean ‘adverb’ and ‘adverbial phrase’.

8 Adpositions include prepositions and postpositions.

9 See figure 18 p.512 in Levinson and Meira (2003) for a complete chart.

10 The scope of x is beside y (or x est à côté de y in French) is more restricted than the scope of x is near y (or x est près de y) and does not cover all the situations in X5. Indeed, the former expressions cannot apply if y is not intrinsically oriented (Peter is near the crowd but not Peter is beside the crowd) or if x is in front of y or behind it.

11 This term subsists in topographic expressions.

12 The fact that most proximity expressions are complex cannot be considered as an argument in favor of their late apparition. Indeed, many simple prepositions in contemporaneous English and French emerge from complex prepositions. This is the case for behind in English (from be + hindran) and derrière in French (from de + rectum).

Haut de page

Table des illustrations

Titre Figure 1
Fichier image/png, 4,2k
Titre Figure 2
Fichier image/png, 4,9k
Titre Figure 3
Fichier image/png, 3,9k
Titre Figure 4
Fichier image/png, 4,8k
Titre Figure 5
Fichier image/png, 11k
Titre Figure 6
Fichier image/png, 7,5k
Titre Figure 7
Fichier image/png, 15k
Titre Figure 8
Fichier image/png, 11k
Fichier image/png, 30k
Titre Figure 11
Fichier image/png, 7,6k
Titre Figure 12
Fichier image/png, 7,9k
Titre Figure 13
Fichier image/png, 15k
Fichier image/png, 34k
Titre Figure 14
Fichier image/png, 39k
Fichier image/png, 35k
Fichier image/png, 34k
Fichier image/png, 29k
Fichier image/png, 30k
Haut de page

Pour citer cet article

Référence électronique

Claude Vandeloise, « The expression of proximity in French and in English »Corela [En ligne], HS-23 | 2017, mis en ligne le 09 novembre 2017, consulté le 10 mai 2021. URL : ; DOI :

Haut de page


Claude Vandeloise

Louisiana State University, Baton Rouge, LA, USA

Articles du même auteur

Haut de page

Droits d’auteur

Licence Creative Commons
Corela – cognition, représentation, langage est mis à disposition selon les termes de la licence Creative Commons Attribution - Pas d’Utilisation Commerciale - Partage dans les Mêmes Conditions 4.0 International.

Haut de page
  • Logo Université de Poitiers
  • Logo MSHS de Poitiers
  • Logo CerLiCO
  • Logo DOAJ - Directory of Open Access Journals
  • OpenEdition Journals
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search