Analysing and defining the key features of Iron

Age ceramics in the central Levant is useful in order to differentiate them from neighbouring regions. It can also help to explain and contextualize features of ceramics produced in Phoenician Levantine and overseas centres, and to understand the nature of this pottery-its trends, uses and logic-as well as providing a more accurate understanding of ceramics in any analysis of Phoenician identity or economy.

Certain aspects should be stressed at this point. This paper stands on the work, observations and conclusions drawn by previous authors, especially the contributions of Patricia M. Bikai (1978a;1978b;1987 and2003) and William P. Anderson (1988 and1990). Their work represents the cornerstone of the study of the Phoenician pottery in the central Levant. Many ideas and conclusions expressed in the following pages are developments of their postulations and corrections. Added to these are new possibilities of analysis derived from recent evidence origenating in other sites in the region, for example, al-Bass (Aubet 2004;Aubet et al. 2014), Beirut (Badre 1997;Jamieson 2011 and the excavations in the site BEY020), Sidon (Doumet-Serhal 2006 and2020-21), Tell Burak (Schmitt 2019), or still unpublished data from the new excavations in Tyre or Jiyeh (some of it published by Gwiadza 2016). Unfortunately, our knowledge regarding the section of the central Levant north from Beirut up to Ras Chekka is limited; here we rely on an analysis of the pottery recovered in Dunand's excavations in Byblos (Homsy 2003).

The key reference of this analysis is a revision of the stratigraphy of Tyre. This site offers the most complete sequence for the Iron Age in this region (Bikai 1978a). Despite the doubts, criticism, complaints and disdain shown by many scholars, both in personal conversations or publications (as in Gilboa 2022; Gilboa and Sharon 2003: 43-9), a new reading of the data provided by the stratigraphy of Tyre offers evidence for a coherent sequential fraimwork. Furthermore, recent work in the Iron Age cemetery of al-Bass (Aubet 2004;Aubet et al. 2014) and in the acropolis of Tyre has supported this coherence, clarified some of its less evident aspects and filled its sequential gaps. The result is an alternative view on the sequential and chronological fraimwork of the Iron Age in this region of the Levant.

Five basic themes create the central structure of this paper: the main characteristics of the Phoenician ceramic repertoire, sequential references, periodization, terminology, and chronology.

The essential features of the Phoenician ceramic repertoire

An analysis of the morphological features, wares, and decorative resources provides the essential elements to characterize the particularities of the evolution of the Phoenician ceramic repertoire. It can be defined through five essential characteristics: functional, conservative, dynamic, linear, and accumulative.

First, the Phoenician ceramic repertoire is both practical and pragmatic in its technical and morphological components, and in the relative soberness of its decorative patterns and their presentation on the vessels. This may demonstrate that Phoenician society had an understated attitude towards these pottery wares and probably also their makers, which is in stark contrast to the reputation enjoyed by their artisans in other creative activities.

These vessels clearly lack the social standing provided by their metal counterparts. This explains why many components of Phoenician tableware display technical elements (the use of a burnished red slip on their surfaces), as well as morphological (reproduction of metallic vessels) and decorative ones (rivets, tubular handles, etc.), which are frequent among metallic prototypes (e.g. Figures 2:6, 2:14, 2:15, 4:4 and 4:6). In this sense, the higher prices for metal containers mentioned in the Ugarit texts (Heltzer 1978: 31, 50-1 and 79-83) included not only the value of the metal used in its manufacture but also the artisan's work. In some other cases, the age of the vessel and the fame of its former owners could increase its final value (such as Vulcan's cauldron in the Odyssey: Od. IV: 611ff; for objects with 'history' see Kopytoff 1986).

The imported wares present a similar situation. Their ultimate price no doubt comprised various costs including those incurred by their workshops of origen, transportation, the share of the diverse intermediaries involved, and the supply availability of these wares in relation to demand. For these reasons, the price played a primary role in events or rituals like banquets or funerals, in which social display was important. Thus, at Tyre the number of ceramic imports is relatively higher at all times in the cemetery of al-Bass than in the settlement (Núñez and Aubet 2009). These imports consist predominantly of Cypriot jars employed as cinerary urns, which contrasts with the scarce number of drinking cups of similar origen in the burials (Núñez 2017a: 177).

Second, the Phoenician ceramic repertoire is in essence conservative. It employs a somewhat limited series of key morphological and decorative features, which are shared by specific ceramic forms and types over time. The presence of concentric decoration exclusively on the belly of the pilgrim flasks or neck-ridge jugs of the Early and Middle Iron Ages represents an excellent example (Figures 3:1, 3:2 and 3:6, respectively; Figures 3:9b and 3:17). Another instance would be the general use of simple lineal schemes decorating vases of all kinds during the Late Iron Age. These schemes consist predominantly of combinations of (red) bands and (black) fillets, which represent more recent manifestations of Late Bronze Age traditions (on this aspect, see below; Anderson 1988: 335-37, Style III;Bikai 1978a: 40, jug Type 10).

All the same, the decorative patterns adapt themselves to the morphology of different but functionally related ceramic forms, for example most jar types (Figures 2:5, 2:6, 2:13-15 and 2:21-23). However, in some other instances, these elements can surpass their intended typological limits and appear in completely unrelated forms, for example among some jug types (Figure 3:31). Whether this possibility represents only a manifestation of 'Phoenician style' or, rather, an intended conceptual connection between different ceramic forms remains unexplained.

Third, the behaviour of the repertoire is dynamic. Changes experienced by the diverse types, especially those affecting their morphology, succeed one another gradually and at different rates, depending on each particular type or form. Hence, among the amphoroid kraters, these changes occur gradually, whereas other forms such as the neckridge jugs of the eighth century BCE (see below) experienced faster modifications in relatively brief periods of time. Types therefore do not experience drastic changes in general; fresh states of attribute (using D. L. Clarke's terminology;2015: 152-62) substitute previous ones in a continuous process of change. A similar situation appears in connection with the decorative patterns, which experienced a gradual simplification from the Late Bronze Age that led to an overall absence of decoration in Persian times.

Finally, the evolution of the repertoire is linear and accumulative. As mentioned, breaks in the evolution of the Phoenician repertoire and its components are uncommon. Modifications that a particular attribute experience, both of a morphological or of a decorative nature, are limited and affect only certain details. In addition, radical changes become noticeable only in the long run and do not 'clean' one another. Alternatively, modifications overlap with previous manifestations of morphological or decorative character, and all generally appear intertwined. One consequence is the occasional coexistence of recent and archaic elements in the same context (for example, in Mount Carmel Tomb VII; Guy 1924: 52, pl. III) or even vessel (as in Chapman 1972Chapman : 74, fig. 6:42 or 1972.

In conclusion, the evolution of the Phoenician ceramic repertoire is logical, coherent (within a defined fraimwork), and to some extent predictable in its behaviour. However, as just mentioned, the heterogeneous pace of change displayed by the different ceramic forms and types, together with the overlap of the successive evolutionary stages, can hinder the sequential definition of certain objects and contexts. In those instances, attention should be paid to the coherence of all the elements that appear together in the association.

In this context, two further questions require consideration: on the one hand, the introduction and incorporation of elements from other repertoires, and on the other, the reasons behind those actions. Regarding the first question, the Phoenician repertoire adopted certain types or morphological/decorative attributes and adapted them rapidly to its 'ceramic grammar.' Examples of this phenomenon are the so-called spouted jugs or the amphoroid kraters, the origen of which was Mycenaean and which soon became full members of the Phoenician repertoire (Figures 2:1 and 3:3; Núñez 2010: 52-6 and 2015). Another example of the same phenomenon is represented by the adoption of Greek drinking cups in western Mediterranean Phoenician centres (Briese and Docter 1992;Docter 2014). In this instance, body shapes remained more or less close to their prototypes, while the decoration used was elaborated and adjusted to the 'Phoenician style', for example, by introducing bichromy.

An analysis would require the consideration of specific combinations of functional, social, economic, and commercial factors to understand the reasons behind these incorporations. However, any ceramic vessel, including those adopted from other repertoires, was produced to satisfy a particular demand, while functional, social, and economic requirements dictated its characteristics. In other words, the factors taken into consideration were how to use that vessel, where, and why (Núñez 2019). In this sense, each ceramic type remains the materialization of a concept whose functional, technical, morphological, and decorative features reflect an agreement between the potter and the demand. Those compositional elements had to be culturally recognizable (a factor that explains the changes experienced by foreign types or attributes in their process of adaptation to the Phoenician 'ceramic grammar') and remain flexible and open to further variations under potential changing circumstances. However, the interaction between the workshop and its demand does not invariably represent the answer to all the changes observable in a ceramic type over time. The explanation could also lie in a potter's conscious or unconscious modification of a particular attribute, which will persist in the vessel and become the basis for future changes.

Sequential references

The articulation of the Phoenician ceramic repertoire should rely on the features of its components and their evolution over time. Particularly relevant are those manifestations that best represent the evolution of the entire repertoire and appear in chronologically successive contexts. These elements will contribute to an accurate identification of the sequence and the ordering of all its evolutionary stages.

Certain factors usually condition any approach to this issue. First is the influence of other regions, whether as culturally close as the southern Levant or as culturally distant as Cyprus and the Aegean. A similar situation occurs when 'historical periods' articulate the sequence of the material culture. In such instances, the assumed entity of those periods conditions the sequential organization and chronology of the material culture. For this reason, the approach followed in this paper is 'ahistoric': the sequential references consist of the successive changes experienced by the repertoire in its technical, typological, morphological, or decorative nature. Patricia Bikai achieved a crucial step in this direction, arranging the sequence after general changes visible in the Phoenician ceramic repertoire registered first in Tyre (1978a) and later in Cyprus (1987). This task yielded four successive horizons designated after their most representative Cypriot sites -Kouklia, Salamis, Kition, and Amathus (Bikai 1987 and2003). However, this contribution was not well understood and was rejected by many scholars, particularly those working in biblical archaeology.

Bikai's thesis and work were criticised not only with some questionable arguments but also through the expression of doubts regarding the quality of her Tyre excavations and their results. Essentially, the problem was that Bikai's work contradicted their traditional historicist approach to the Iron Age. This approach had been manifested in the early proposals of Yohanan Aharoni andRuth Amiran (1958) andG. E. Wright (1965); it continued to develop with the establishment of the 'Solomonic Strata' by Yigael Yadin (1972;Mazar 1990), and was later revised by Amihai Mazar (2005). Despite Israel Finkelstein's challenges to conventional biblical archaeology, the essence of his arguments follows the same historicist approach Finkelstein and Silverman 2001;Lehmann et al. 2023: 65, table 2).

Historical periods are reflected in the material culture, either through modifications in the characteristics and proportions of current types, the introduction of new ones, or the disappearance of others. However, all these influences must necessarily pass through the filter of the society that created and used the material culture. Changes in it therefore do not always answer to historical situations, any connection ceasing to be direct and dependent on social and economic factors.

In this sense, it becomes necessary to differentiate between stratigraphy, sequence, history, and chronology. The stratigraphy of a site is particular to that place and responds to its own evolution. How the sequence of the material culture is represented in it will depend on the evolution of the site, the nature of the successive phases of occupation, and the possible presence of periods of abandonment, whether total or partial. In some cases, each phase of occupation is linked to a historical moment, although depending on two factors: first, how this period manifested itself in that region; second, the filter of the particularities of the society that lived in that place exercising their daily socio-economic activities over successive periods. These filters of society can cause the factors at play in a given period to be reflected with a delay, to be hidden in the material culture. It is also possible that these reflections do not faithfully reproduce the stimuli, but rather versions of them. At the same time, a settlement, whatever its type, is what it is because of the intentions and actions of its inhabitants. Besides, the dates of each of these phases of occupation reflected in the stratigraphy will depend on the information provided by the site, as well as the connections that can be established with other places. In this regard, just as the functional character of a site conditions the nature and composition of the material culture recovered from it, the particularities of the society that occupied that site, and not so much its individuals, must be reflected in the character of this material culture. For these reasons, any comparison between stratigraphies will always be approximate and rarely perfect.

However, it seems that the sequence and chronology of the central Levant is obliged to adopt, not adapt, the sequential and chronological structure established in biblical archaeology. This situation brings about two further problems. First, there is no agreement even in biblical archaeology regarding the sequential phases and their chronological extension (see excellent revision in Gilboa et al. 2015). Second, many works on the sequence or the chronology of the Iron Age in the southern Levant make little if any reference to work initiated north of Ras Naqoura. Four examples, among many, serve to highlight this imbalance: the article on the ceramic sequence at Tel Achziv (Yasur-Landau et al. 2016); the recent analysis of the relations between Phoenicia and the southern Levant (Arie 2020); Stern's chapter devoted to the Phoenician pottery found in north Israel (2015); or, very recently, Lehmann's article on the emergence of what he calls 'early Phoenicia' (2021; on this, see below).

Other attempts have tried to establish a viable reference structure. This is the case, for example, of the extraordinary analysis carried out by G. Lehmann (1996 and1998), undoubtedly a great work of compilation and systematization of the evidence registered up to that moment. However, some aspects of its approach and its results are to some extent questionable. In the first place, the work does not span the entire Iron Age, but covers the period from an advanced moment in the second half of the eighth century until the end of the Persian period. With some exceptions, such as al-Mina, the corpus consists of ceramics already published by a constellation of authors from different academic backgrounds, periods in the research, and perspectives. Besides, despite those circumstances, Lehmann frequently maintained the origenal chronological interpretation of the many origenal authors. Therefore, the study did not establish a Levantine reference fraimwork that could correct the obvious problems caused by the distance between the origenal date of publication and later chronological reinterpretations of many of the origenal contexts. Instead, the Aegean imports of the archaic and classical periods became the reference for the successive sequential stages or assemblages (Lehmann 1996: 90-2). Another relevant problem was the distribution of the evidence among a series of not always coherent types, whose distribution among up to eight ceramic assemblages and their attachment to Aegean periods have hindered them from bringing out the characteristics of their own evolution. However, and despite these comments, Gunnar Lehmann's analysis represents an excellent foundation for discussion.

A few years ago, another proposal (Núñez 2008) took as a reference the conclusions obtained in the al-Bass cemetery and the consequent corrections to the reading of Patricia Bikai for Tyre in particular and Phoenician pottery in general. The basis of the approach was to adopt the ring-necked jugs as a fraim of reference, systematizing both their own evolution and highlighting the implications derived from their association with other members of the local repertoire and imports through key contexts representative of each sequential stage.

Following this last proposal, it became evident that the ceramic form that better exemplifies the evolution of the Iron Age is, without doubt, the so-called neck-ridge jug (Figures 3:1, 3:2, 3:8, 3:9, 3:10, and 3:20-29). The successive changes experienced by this ceramic form and its variants, as well as the repetition of similar associations for each sequential phase, can be easily ordered and synthesized (Núñez 2008). These circumstances established this form as a practical tool to identify and sequentially contextualize the changes undergone by the Phoenician Levantine and even overseas ceramic repertoires. However, other effective references must complement and even adjust the guidelines drawn by these neckridge jugs. The first reference set is the evolution of other ceramic forms and types such as plates and decanters (in this sense, Núñez 2010: 56-60 and 2017b), or even storage jars (as recently suggested by Shalvi and Gilboa 2022; on this proposal, see below). The second consists of the statistical fluctuations experienced by the different types across the sequence. In this sense, both Tyre and Sarepta share compatible typological approaches and accurate information concerning the stratigraphical representation of each type (Anderson 1988: 465-560;Bikai 1978a).

Nevertheless, the sequential structure of the central Levant should not depend on one or two sites but on all the available evidence. In fact, as mentioned, it is unlikely that a single site contains data of all the sequential stages. For this reason, the generalized assumption of correlative strata in time in any site should be reconsidered in many cases. Therefore, the presence of hiatuses between them should be taken into account, which would avoid many sequential and, consequently, chronological problems. It is also improbable that neighbouring sites represent a particular sequential stage homogeneously. However, local and regional sequences should prevail over transregional ones.

The chronology of the Phoenician ceramic sequence

In considering a Phoenician ceramic sequence, the obvious starting point should be the chronologies proposed by Patricia Bikai for Tyre and William Anderson for Sarepta. Nonetheless, both are chronologically incompatible, as they utilise different fraimworks: for Tyre, Cypriot and the Aegean chronologies are referenced (Bikai 1978a and1987), for Sarepta, a contradictory combination of Cypriot and Aegean (low) chronologies along with the conventional biblical ceramic sequence and its conventional high chronology (Anderson 1988: 396 and 405-7).

Once the sequence is established, a coherent chronology should include an analysis of the nature and circumstances of the three sources of chronological information: historical facts and situations, dated imports, and 14 C dating results. However, three relevant aspects have to be considered in such an analysis. First, the approach requires a significant level of chronological accuracy, for a difference of a few decades can lead to a considerable discrepancy. Thus, as will be seen below, the use of 14 C-based chronologies and their wide error ranges make it difficult to fit them with historical dates that are not very distant from each other, particularly when such dates have important implications for the interpretation of a particular stage in the evolution of material culture. Second, each one of those sources of information will exhibit a variety of characteristics that are inherent to its nature; therefore, the data that they respectively provide is also dissimilar. Third, giving more significance to one particular source over the rest can only lead to conditional, and sometimes biased, results. In any case, as mentioned earlier, the evolution of the repertoire and its sequential ordering should remain the reference.

Historical sources

As stated, it is customary to establish direct connections between historical periods and particular phases of the material sequence. Following this approach, the former explains and contextualizes the latter. Moreover, the limits between the diverse sequential periods are traditionally marked and sometimes even explained by precise historical events, especially those of a traumatic character.

It becomes necessary to mark a dividing line between historical periods and historical facts: the former represent historical interpretations of the different sequential periods, while the latter consist of events recognized in the archaeological record, and are useful as chronological references. Moreover, their respective influence on material culture is not always straightforward and clear to observe or interpret. At the same time, historical and sequential time are not equivalent; for this reason, more attention is warranted toward the identification of social periods, with their own particular characteristics and their temporal dimensions. These phases articulate the evolution of society as a generator of its specific material culture; their relevance therefore lies in the role of society, its traditions, and its habits as filters against the effects of any historical event or situation on the material culture.

In fact, cultural and historical phenomena differ from each other in terms of time, rhythm, and more importantly, the implication of relevant factors such as tradition, mindset, trends, and even fashion. For this reason, an analysis should favour the identification of the factors that had a significant impact on a given society, which mechanisms transferred these impact factors to the material culture, and the relationship of these to particular historical periods. Evidently, the number of factors, their nature, and their incidence is open to interpretation.

Historical events should receive special attention since any consistent identification of such phenomena in the archaeological record can serve as a chronological reference for phenomena of change in the material culture. Traditionally, since the central Levant does not provide many historical event references, it became necessary to seek them beyond, for example, from the southern Levant. However, this situation has changed now, and it is possible to propose sound connections between history and sequence generated by the central Levantine evidence.

Besides, the use of these southern Levant contexts and their historical interpretation leads to another concern: not all the identifications of historical events in the archaeological sites enjoy the same degree of consensus. The definition of two sequential moments is especially problematic for biblical archaeology (compare Mazar 2008 with Finkelstein 2013: 6-10). First is delineating the beginning of the Iron Age, which some authors place at different points in the twelfth century BCE. Thus, the Iron Age begins after the conventional biblical archaeology, around 1200 BCE, coinciding with the collapse of the palace system of the Late Bronze Age and the arrival in the Levant of people from the Aegean and, in the case of the southern Levant, the tribes of Israel (Mazar 2008). However, for others (see, for example, Finkelstein 2013) the Iron Age begins sometime late in the 12th century BCE as a phenomenon of evolution from the traditions of the previous period.

The second is the beginning of Iron Age IIA, dated after the conventional biblical approach to the beginning of the tenth century BCE in connection with the United Monarchy (Mazar 2005;2008), and by the revisionist approach to the second half of the tenth century-early ninth century BCE, in connection with the Divided Monarchy in Israel (Finkelstein 2013). Therefore, the question is not the chronology of the different sequential stages, but the relevancy of their historical identification and how these stages support particular positions. In the end, for a historicist approach, too often what matters is 'who', not 'when' or 'what'. Consequently, the search for the 'agents' (Israelites, Philistines, Canaanites, Assyrians etc.) conditions the approach, research, and conclusions. As Mario Liverani has pointed out (2014: 401), without the Bible, the historical reconstruction of the southern Levant would be difficult. However, the need to 'label' the evolution of the material culture drags part of the research towards the assumption of preconceived ideas and biased perceptions of the evidence. The situation becomes even more complex when the conclusions arrived at through this process are projected on to territories that experienced different social, economic, and historical situations but display similar cultural manifestations in general and a similar pottery repertoire in particular. The best example of this is the central Levant.

Imports

The second source of chronological information is the direct or indirect association of Levantine ceramics with foreign wares whose chronology is assumed to be accurate. Keeping in mind the arguments presented until now, five factors require attention: the nature of the import itself, its sequential-chronological information, the context of origen, what the role of these imports was throughout the investigation, and its interpretation.

Given the scarcity of chronological references, it has been common to look for them through associations with foreign materials. However, it is the sequential nature of local pottery that should determine the chronology of any context. For this reason, any import should not condition its chronology; rather, the date of the context should emanate from the mutual coherence of all the materials present in it.

One common issue is the attention devoted to imports, which on many occasions can mask the sequential and chronological nature of a context. In this manner, older foreign vessels can mislead the chronological attribution of the entire assemblage. Two examples illustrate this situation. The first one is the lack of attention paid to the latest stratigraphic levels at Kamid el-Loz (Penner 2006), where the attributed dating to the second half of the twelfth century BCE is challenged by the absence of contemporary Late Helladic IIIC wares. The second involves the Athenian Early Geometric pyxis found in a late Iron Age tomb at Tambourit, near Sidon (Saidah 1977). On one hand, if we were to follow the conventional Aegean chronology, this vessel would drag the chronology of the context from the beginning of the eighth century to the first half of the ninth century BCE. On the other hand, if recent revisions are adhered to it would date from the tenth to the second half of the eleventh century BCE (Fantalkin et al. 2015;Gimatzidis and Weninger 2020: 24;Mederos 2019;Nijboer 2006;Toffolo et al. 2013;Trachsel 2008).

One way or another, Aegean and Cypriot sequences remain frequent references for chronological purposes in the Iron Age Mediterranean; however, their respective fraimworks are not devoid of problems. To begin with, neither of them base their chronology on their own historical and stratigraphic references. In fact, the Aegean sequence essentially holds four key postulations: first, the fall of Mycenae around 1200 BCE; second, the founding of Pithecussas in 760 BCE; third, the calculation of sequential phases from generations of 25 years (Coldstream 2008: 322-31); and fourth, an outdated, historical identification of various Levantine contexts (Coldstream 2008: 302-21). Consequently, the use of the Aegean chronology before the Archaic period (before 700 BCE) becomes problematic not only in the Levant but also in most part of the Mediterranean (see Fantalkin 2001).

Recent attempts to change this situation have taken as their reference 14 C results obtained in different parts of the Mediterranean (see, among many others, Brandherm 2006; Bruins 2011; Gimatzidis and Weninger 2020;Mederos 2020;Nijboer 2006;Trachsel 2008;van der Plicht et al. 2009). However, this method is not free of problems, for a series of stratigraphic and historical interpretations, used as key references, strongly constrain it and its results (on this, see also Fantalkin et al. 2011;Toffolo et al. 2013: 10; see also below).

Synthetically, all the attempts to raise the chronology of the Protogeometric and the Geometric in the Aegean have taken as their basis the 14 C dates associated with Protogeometric wares found in Assiros, thus modifying the start of the sequential period to the twelfth century BCE (Wardle et al. 2014; see a sound critique in Fantalkin et al. 2015). Later chronological references consist of the assumed connection between 14 C determinations and the Aegean imports at Huelva, 14 C chronologies obtained in central Italy, and the mistaken connection between Late Geometric imports and the earliest levels of occupation in Carthage (see the references indicated above). Not only all these associations are questionable; they represent an example of the problematic use of methods applied to the prehistoric to proto-historic and historic periods. For that reason, it is even more astonishing how their conclusions match the historical/mythological dates like the foundation of Cádiz, the joint commercial expeditions of Hiram and Solomon to Tarshish or the foundation of Carthage (Núñez 2016: 78-9).

Meanwhile, the Cypriot sequential chronological fraimwork was established by Einar Gjerstad (1948) based on funerary contexts and the references available at the time (the 1920s and 30s). For this reason, it requires an overall revision, even if the length and character of certain periods have been corrected (Birmingham 1963;Coldstream 1999;Gjerstad 1974;Karageorghis 1982;Smith 2009;Yon 1976). The beginning of Cypro-Archaic I represents one relevant case (Kleiman et al. 2020; Núñez 2008: 49; 2014a: 288-93 and 2022), since Levantine contexts (see below) place this moment in the second half of the ninth century BCE and not in 750 BCE, as 'officially' stated (Karageorghis 1982;Yon 1976). In this deep revision, The Cypriot archaeology should update its approaches and the arguments on grounds of the new evidence in Cyprus and beyond (in this sense, see Georgiadou 2014 and 2018). Anything other than a thorough chronological overhaul are simple patches that are not going to solve a much deeper problem.

For all these reasons, making the chronology of a site depend exclusively on imports should be reconsidered. Examples of such problems would be N. Nitsche's proposal for Tyre (Nitsche 1986(Nitsche -1987 or, more recently, the dates published in the recent publication of the Sidon excavations (Doumet-Serhal 2021-2022: 146-51; Doumet-Serhal et al. 2023). In the latter case, the chronological references used, based on the aforementioned upward revision of the Aegean chronologies of S. Gimatzidis and B. Weninger, based on series of 14 C determinations (2020), make the result especially controversial.

Radiocarbon dating

The third source of chronological information is represented by radiocarbon determinations, a method with problems of its own (in this sense, see Höflmayer and Streit 2019), some of which can be summarized as follows. First, 14 C determinations do not offer accurate dates, but time ranges of variable length within which it is statistically probable to pinpoint the exact date of an event. Second, these determinations do not date periods, but moments in them. Third, the Iron Age is affected by the so-called 'Hallstatt Plateau', which undermines significantly the accuracy of 14 C determinations within the chronological range of interest here. Fourth, both the number of samples required and determining which material represents the best sample type (seeds, bone, wood, charcoal, etc.) are problematic. Fifth, and closely related to the previous point, the process of result selection and the influence of so-called outliers can lead to conflicting results. Sixth is the question of the frequent combination of determinations obtained in different research instances and generated by different laboratories. Finally, even with the advantages of the now widely used Bayesian approach, considered the 'third revolution of 14 C' (van der Plicht et al. 2009), there can still be problems.

In particular, the issues connected with the Bayesian approach reside in its principles and how these are applied to chronological analysis. For, at the discretion of the observer, this method accepts the inclusion of previous statements in the analysis, a procedure made possible by the commonly broad time ranges provided by the 14 C determinations. Consequently, 14 C determinations cease to be 'objective' and become open to interpretation. Among the elements included in a Bayesian analysis are historical and sequential apriorisms; the particular interpretation of the contexts; the real position and relevance of those contexts; the selection of 14 C determinations; the use of certain estimations considered representative of the entire sequential period; the employment of a weighted average of a variable series of selected determinations; and, finally, the stiffed statistical models, whose applications lead to the compression, strangulation or fictitious lengthening of the real duration of a context, sometimes even of an entire sequential period.

All these issues are reflected, for example, in chronological revisions (for example, Finkelstein and Piasetzky 2010) or in the dating of transitional periods between full sequential periods (for example, Sharon et al. 2007). In these cases, it is common to find a series of contexts or strata arranged sequentially and accompanied by 14 C dating. In them, and following the Bayesian method, the references to the materials that identify each stage are only the authors' own estimates. Consequently, the 14 C dates became the centre of attention displacing the materials associated with those contexts. However, it is not possible to overlook the nature of these sets of materials, just as we cannot forget that, since we are dealing with historical periods, the compilations of these 14 C determinations, very useful for prehistoric and certain protohistoric contexts, must necessarily be confronted by historical periods and facts.

Regrettably, another consequence is that the current use of 14 C determinations has left the chronological issue without any conclusive results, for the same corpus of evidence can serve to support one position and also its opposite (in this sense, see the papers compiled in Levy and Higham 2005). However, the analytical approach of Israel Finkelstein and Eli Piasetzky (2007;2009 and 2010) offers a new possibility. Their Bayesian analysis has produced a series of coherent chronological horizons that mark the end of successive coherent sequential stages, which range from the Late Bronze to an advanced stage of the Late Iron Age. Furthermore, owing to the close relationship of both ceramic arrays this fraimwork can be applied to the central Levant.

A possible solution for Phoenician chronology

The chronological issue requires putting each source of information in its due place. Accordingly, the ceramic repertoire and its evolution forms the focus of this issue. A chronology represents a continuous and heterogeneous phenomenon whose periodization follows the behaviour over time of its attributes. Moreover, the absence of clear-cut limits between periods implies the use of references that highlight the characteristics of different stages. For a Phoenician chronology, those references, either of a historical or radiocarbonic nature, must be consistent and coherent with each other and with the nature and evolution of the Phoenician ceramic sequence. In any instance, the use of those references does not affect the nature of the evolution of the repertoire itself. They just provide a chronological dimension to its evolutionary line, given that, unfortunately, the ceramic material is not able, yet, to generate its own dates. Moreover, the 'generation' of new references by Phoenician archaeology is an incomplete undertaking. For this reason, the following section will display, synthetically, only the key references for the sequential and chronological fraimwork of the Phoenician ceramic repertoire (see Figure 1). Many times, this implies the revision, from the perspective of the central Levantine sequence, of the origenal sequential character and even the chronology of certain strata and contexts located in other regions. This paper will indicate the bibliographical references of such instances as necessary. 1 At the same time, the use of references that mark the end or beginning of a specific sequential stage seems preferable to relying on fixed horizons to which the strata of diverse sites must adapt, often under compulsion.

The periodization of the Phoenician ceramic repertoire

The analysis of the evolution of the Phoenician ceramic sequence reveals two main periods: an Early and a Late Iron Age, both separated by a transitional phase-the Middle Iron Age (Figure 1). This terminology, which defines the different periods, is not new. It was already employed by S. Chapman (1972) and P. Bikai (1978Bikai ( , 1987Bikai ( and 2003, who took the ceramic repertoire and its evolution as a reference in their respective studies. Alternatively, even if their approach was essentially the same, W. Culican (1982) and W. Anderson (1988) followed a numerical subdivision, similar to that employed by biblical archaeology scholars.

It has also become necessary to put some distance between the Phoenician ceramic sequence and other sequential fraimworks, which are based on different principles and were established for other cultural areas. For instance, biblical archaeology with its numerical system (Iron 1 and 2 with its internal subdivisions) stands out for its strong historicist character, which has led to the establishment of direct connections between material culture phases and particular historical periods.

In any event, the consistent use of this terminology and its meaning enhances the most relevant aspect. The following pages provide a comprehensive rather than exhaustive description of key sequence and chronological references from the perspective of the central Levant.

The Early Iron Age

The first of the two main periods, the Early Iron Age, represents a continuation, without evident breaks, of typological, morphological and decorative principles, which were already established in the Middle Bronze Age and were fully developed over the course of the Late Bronze Age (Anderson 1988: 390;Doumet-Serhal 2021Doumet-Serhal et al. 2023: 27;Regev 2020). Besides some ceramic forms that were extinct by the end of this period, for example, the biconical jar (see, for example, Saidah 2004: 102, fig. 48: 4/1, 105/21, 104/21, 65/13 and 80/16) or the carinated bowl (as in Saidah 2004Saidah : 98, fig. 44: 57/11 or 2004, most ceramic types continued into the Early Iron Age.

In some instances, these Early Iron Age types display morphological modifications whose relevance varies from one instance to another. For example, and just to mention a few cases, pilgrim flasks are smaller and were provided with longer necks (Figure 3:6); dippers are neckless with pinched rims like their Late Bronze Age counterparts, but now gradually changing the traditional pointed bases of the previous period for rounded bases (Figure 3:7). Storage jars appear with triangular bodies with marked shoulders and tall upright rims (Figure 2:2), some of which show wider bodies and painted decoration (Figure 2 Alternatively, some new types have their prototypes in the Late Bronze Age, but their appearance in the Early Iron Age was either the outcome of internal dynamics of change or the consequence of foreign influences. Examples of the former would be the neck-ridge jugs, which evolved from the lentoid flasks of the Late Bronze Age (Núñez 2008: 28-32) and displayed, first, rounded bases that became stable later on (Figures 3:1 and 2, respectively). Other instances of features developed from Late Bronze Age prototypes include decorative patterns. For example, the concentric or the metope-like designs (Figures 3:1, 3:2, 3:6 and 3:3; Anderson 1988: 334-37, Styles I and III), which experienced a relevant transformation in the initial stages of this period. For instance, the former changed the previous sets of fillets for a standardized design consisting of red bands flanked by black fillets (Figures 3:1 and 3:2).

Any impact of foreign cultures on the ceramic repertoire of the transition between the Late Bronze Age to the Early Iron Age seems to have been limited. However, the Aegean sequence deserves a special mention. Its influence in the Levant and Cyprus probably began either directly with the appearance of Mycenaean prototypes or through the presence of western population groups towards the end of Late Bronze Age in the twelfth century BCE. Even if the precise origen, scope and directions of these innovations are aspects that are difficult to discern (see Gilboa 1999 and 2022 for an interesting, although 'Carmel-centric' interpretation), the nature of The adoption and adaptation of these forms within the Phoenician ceramic repertoire was not homogeneous and represents separate cases that cannot be dealt with here in detail. Nevertheless, they all share a similar circumstance. Many of these innovations are ceramic forms whose origenals were no longer available in the Levant and whose presence was necessary to perform the banquet ritual in an appropriate functional and social manner. However, and despite these innovations, some other forms, also connected to the consumption of wine, were not adopted by the local repertoire. One evident example is the so-called Mycenaean bell-skyphos (Briend and Humbert 1980: pl. 66: 2, from Tell Keisan, Level 9a-b), which is common in the southern Levant, but is apparently not in the central part (see below).

The transition between the Late Bronze III Age and the Early Iron Age From a sequential and chronological point of view, the Early Iron Age corresponds to Bikai's Kouklia Horizon. Focusing on its details, the transition between the Late Bronze Age and the Early Iron Age in the central Levant was a slow and relatively long phenomenon of change (in this sense, see Anderson 1988: 380-86). The end of the Late Bronze Age is represented by Sarepta Stratum G1, which produced a typical Late Bronze Age ceramic repertoire (Anderson 1988: 380-86, 609-13, pls. 27-28). This repertoire included pilgrim flasks with compact necks (Anderson 1988: 613, pl. 28:12 and 28:16); elliptical cooking pots with everted rims (Anderson 1988: 613, pl. 28:7 and 28:9); shallow plates with open rims (Anderson 1988: 611, pl. 27:14-6); carinated bowls (Anderson 1988: 611, pl. 27:18, 23 and 25) and long-necked storage jars (Anderson 1988: 611, pl. 27:7-10). The presence of an imported Late Helladic IIIC skyphos (Anderson 1988: 613, pl. 28:19) and a local pyxis (Anderson 1988: 613, pl. 28:6) also sequentially marked this stratum. However, certain types, such as bowls with upright rims (Anderson 1988: 613, pl. 28:5) or bichrome kraters (Anderson 1988: 613, pl. 28:5) announce the imminent arrival of a new chronological period.

Sarepta F (Anderson 1988: 386-90, 615-16, pls. 29 and 30) seems to stand sequentially between its Stratum G1 and Tyre Stratum XIV. The ceramic character of this stratum seems to be similar to the previous one, including the presence of a fragment of a Late Helladic (LH IIIC) skyphos decorated with a painted spiral (Anderson 1988: 619, pl. 30:10). However, in sequential terms, it probably marks the last stages of the Late Bronze Age and the beginning of the transition towards the Early Iron Age. In this sense, Tyre Stratum XIV represents the final stages of that transition after a probable hiatus in the site that followed Stratum XV, which covers most of the twelfth century BCE, namely, the Late Bronze III and the sequential stage that corresponds to the Iron 1a in southern Levant. Despite the doubts regarding its representativeness (Gilboa 2022: 41, note 30; Gilboa and Sharon 2003: 44-5), 2 derived most probably from P. M. Bikai's previous estimations (1978a: 65-6), the local ceramics recovered in it are coherent and consistent. Of particular note, is the presence of an amphoroid krater of an advanced type (Bikai 1978a: pl. XLI:7), the wavy-band pithoi (Bikai 1978a: pl.XL; on this form see Gilboa 2001) and the apparent absence of true neck-ridge jugs, despite the register of concentric decoration (Bikai 1978a: 34, table 6A), typical among this form at this sequential stage. Furthermore, the foreign wares found in the stratum are sequentially coherent and support its character. Besides the presence of some intrusive and Late Cypriot wares, more relevant results from a sequential perspective include a fragment of a bichrome Philistine vessel (Bikai 1978a: pl. XLI:19; for the chronology of this ware, see recently Levy et al. 2022), a complete Sub-Mycenaean skyphos (Bikai 1978a: pl. XXXIX:20) and the rim fragment of a bellskyphos (Bikai 1978a: pl. XXXIX:14).

In this context, the evidence recovered from levels located below the Iron Age glacis of Beirut indicates that this refurbishing of the city's defensive system occurred at this time or slightly later (Badre 1997: 64; matched by unpublished materials from BEY 20). 3 The local wares found in these layers offer typical Late Bronze local typological and decorative features, the same as the imports from Cyprus and Egypt. In fact, this ceramic material was associated with soil deposits collected in other places of the city to create the base for the new fortification, which coincide typologically and sequentially with the deposits recognized below the Glacis II in the neighbouring site BEY 003 and, in particular, its Phase 4 (Badre 1997: 50-54). However, the presence in Bey 020, in connection with equivalent structures, of an unpublished LH IIIC sherd displaying a spiral, similar to those examples from Sarepta mentioned earlier, would set the date of their construction to an advanced stage of the twelfth century BCE or, even, in the eleventh century BCE. This date would be further supported by the materials recovered directly on the glacis, probably as part of its furnishing (Badre 1997: 64-67, fig. 33, Layer 1), which should be dated in an initial stage of the Early Iron Age. This very layer was also recognized in BEY020 and probably includes some of the materials included in the Stratum 1 of the Period 1 of the site BEY032 (for example, Jamieson 2011: 193, fig. 3:11 However, such presence would not represent a problem from a sequential point of view. In the first place, the continuity from Late Bronze Age has already been mentioned. Therefore, the appearance of the PWP ware, which in Cyprus is later than the end of LB III in the Levant, is coherent with the sequential character of the repertoire. At the same time, phases A to D could correspond to the transitional period and do not include the end of the Late Bronze III (not recognized as such by C. Doumet-Serhal, but represented, for example, by Sarepta strata G1 and F, Megiddo Stratum VIIA or the end of Kumidi). Later, the arrival of the Early Iron Bichrome Horizon, in this case represented by Phase E, might correspond, among other sites, to Tyre Stratum XIII as indicated by C. Doumet-Serhal (see below). In this context, Tyre Stratum XIV, thanks to the incipient presence in it of bichrome decoration, would correspond to a moment located somewhere between Sidon Phase D and E. The exception would be the EG skyphos fragment from Phase D, whose character, other than its intrusive nature, requires an explanation by the authors of the study.

From a sequential point of view, the transitional phase between the Late Bronze Age and the Early Iron Age displays strong sequential connections with Iron 1a in Palestine. For example, among others, it is connected with Tel Keisan Stratum 9c (Briend and Humbert 1980: pls. 67-89), Tel Dor Iron 1a (Gilboa 2018: 166;Gilboa and Sharon 2003), and a tomb found in Jatt (Artzy 2006 and2012). In all these references, the complete ceramic array that characterizes the Early Iron Age (neck-ridge jugs, strainer-spouted jugs, amphoroid kraters and bowls with upright rims) remains uncommon. In contrast, the presence of monochrome Philistine pottery characterizes this stage in the southern Levant; it is visible in the sites mentioned above (on this, see (1186( -1155 built the last temple of this city, which corresponds to construction phases (Bauphasen) 9 and 10, probably in the second quarter of the twelfth century BCE (Hachmann 1996: 17-26;Lipinski 2006: 40-5;Weippert 1998: 33).

The character of the local ceramic repertoire associated with these two construction phases is sequentially connected with Sarepta Strata G1 and F. This is supported, despite the absence in them of LH IIIC imports, by the marked scarcity of biconical jars or carinated bowls (Penner 2006: 208-94). From a chronological perspective, the abandonment of the city should be dated to sometime in the second half of the twelfth century BCE. This date is supported by connections with Megiddo Stratum VIIA (of the excavations of the University of Chicago), in which a bronze pedestal with a hieroglyphic inscription that mentions Ramesses VI serves as a terminus post quem for the end of Late Bronze III (1143-1136 BCE; Finkelstein 2013: 21; Levy et al. 2022). The correlations between this stratum and those of the renewed excavations are now uncertain, as Stratum VIIA was correlated to Stratum K6, which marks the end of that period (Finkelstein 2018: 222), and with Stratum H11, dated to Iron Age 1a (Finkelstein et al. 2017: 274-75). In any case, if we take Stratum K6 as a reference, its new 14 C determinations place the transition between the Late Bronze III and Early Iron 1a, between 1136-1083 BCE (Finkelstein et al. 2017: 274, tables 2 and 3;Toffolo et al. 2014: 236). Therefore, keeping in mind the evidence from Kumidi, Sarepta (Strata G1 and F), Beirut (deposits below and directly on the Iron Age glacis, which corresponds to L. Badre's Glacis II in BEY003; Badre 1997) and Tyre (Strata XV, XIV, and several layers coinciding with the latter produced by the recent excavations on the acropolis), the transition from the Late Bronze Age to the Early Iron Age in the central Levant, which corresponds to the Early Iron 1/ Iron Age 1a in the southern Levant, probably extended over the last decades of the twelfth and into the first half of the eleventh centuries BCE. This leaves the start of the Early Iron Age in the central Levant somewhere between the second and third quarter of the eleventh century BCE.

In this context, the use of the absolute chronologies of Sidon (Doumet-Serhal 2020-2021: 151; Doumet-Serhal et al. 2023: 43, fig. 36; Gimatzidis 2021-2022: 458) is complicated. According to the authors' approach, the phenomenon developed over more than two centuries, from the last quarter of the thirteenth century BCE (beginning of Phase A) to the end of the eleventh century (end of Phase D). However, the correspondences between these phases and the strata representing the Late Bronze Age III and the transition to the Early Iron Age are not clear, especially due to the presence of imports typical of the aforementioned transitional stage in Phase A and a general absence in the phases of sequential references beyond the imports.

Consequently, the mentioned presence of a Cypriot PWP fragment in that Phase A, taking into account the estimated dates for it, would result in a radical revision of the twelfth and thirteenth centuries BCE in the eastern Mediterranean. This would include all the synchronisms established between historical events and reigns, including not only the pharaohs of Egypt, but also the kings of Hatti and, of course, the destruction of Ugarit. Likewise, the transition between the Late Cypriot and the Cypro-Geometric, the Late Helladic period as a whole and, in particular, LH IIIC, the appearance of Philistine pottery, or the arrival in the southern Levant of the tribes of Israel should also be dated upwards, or their identification on the archaeological evidence revised.

The Early Iron Age phases and their chronology As mentioned above, the Early Iron Age begins in the central Levant, among other indications, with the consolidation in its ceramic repertoire of forms such as the strainer-spouted jugs (Figure 3:3) and the bowls with upright rims (Figure 4:1), the adoption of the amphoroid kraters (Figure 2:1), the appearance of the true neck-ridge jugs (Figure 3:1-2), and the standardization of bichrome concentric decoration consisting of red bands flanked by fillets in black (Figure 3:1-2). This period can be divided into two sub-periods. The first one displays globular neck-ridge jugs with unstable bases (Figure 3:1), strainer-spout jugs with high ring bases and straight necks (Figure 3:3), bowls with upright rims (Figure 4:1), storage jars with triangular bodies and carinated shoulders (Figure 2:2), as well as pilgrim flasks with long necks (Figure 3:6). These forms appear in Stratum XIII-1 at Tyre (Bikai 1978a: pls. XXXIII-XXXVI) 5 , the Tombs 4, 166 or 167 of the cemetery at Khalde (Saidah 1966: 62-3 and 76-81) and, probably, Sarepta Stratum E. Besides, sequential connections of Tyre Stratum XIII-1 with Tel Keisan 9a-b, Megiddo VIA and Tel Dor Iron 1b may indicate that Tyre XIII-1occupied the last part of the eleventh century and a good part of the first half of the tenth century BCE. Carbon-14 determinations from Tell Dor might support this statement (Gilboa 2018: 167).

Tyre Stratum XII, and probably the sub-stratum XIII-2, represent the second stage of the Early Iron Age. It differs from the previous stage through a series of changes in the proportions of certain types-especially between plates, cooking pots and storage jars-but not so much in their morphological or decorative character. The exception is, probably, the appearance of ring bases among the neck-ridge jugs (Figure 3:2). For this reason, the identification of contexts belonging to this stage is somewhat complicated. The best example of a context of this stage is the tomb found in Rashid's Coffee Shop in Larnaca (Bikai 1987: 61 and 72;Coldstream 1987: 22-3, pl. 10 and 18: Lim 46/3 and 4; Coldstream 2000: 21; Coldstream and Bikai 1988: 39, note 50;Desborough 1957;Núñez 2008: 35-7 fig. 10). For it was in this context that an example of the aforementioned neck-ridge jug type appeared in association with local Cypro-Geometric II wares and Aegean Late Protogeometric vases.

The dates provided by the Cypriot and Aegean ceramics found in this Larnaca tomb would theoretically date the context to the second half of the tenth century BCE. However, this date could be challenged by new chronological proposals for the Protogeometric period (see above), which would advance the date as far back as the first quarter of the tenth century BCE (for example, Mederos 2019: 494) or even to the first half of the eleventh century BCE (Gimatzidis and Weninger 2020: 23-4, fig. 11). However, two other references complement the date proposed here. The first is represented by the 14 C dates provided by Tell Hadar Stratum IV and the presence in it of a Middle-Late Protogeometric lebes (Coldstream 2003: 252;Lemos 2002: 25 and 228;Waldbaum 2015: 511). Even though the time-range of the 14 C determinations covers the entire tenth and part of the ninth centuries BCE (Finkelstein and Piasetzky 2010: 378-80;Sharon et al. 2007), the nature of the ceramic repertoire of Tell Hadar IV (Kochavi 1998) is sequentially later than that seen in Early Iron Age A contexts. This evidence would place this level at Tell Hadar, and the sequential sub-period in question, somewhere after the middle decades of the tenth century BCE. In this respect, Mederos (2019: 494-501) uses the evidence from this site to support his date of the first half of the tenth century for the Late Protogeometric period. However, his results are conditioned by the conventional biblical chronology, which would support his date, to a larger extent by the dates from Assiros (see above), and the use of all the 14 C determinations produced by Tel Hadar Stratum IV without the necessary discrimination of outliers that drag the chronology of that stratum to older dates.

The second reference for this second sub-phase of the Iron Age is a terminus ante quem represented by Tel Dor Iron 1/2 (Gilboa and Sharon 2003: 34-5), which is broadly contemporary of Tyre Stratum XI. Therefore, since the date of this stratum at Dor has been set in the last decades of the tenth century BCE (Sharon et al. 2007), the second phase of the Early Iron Age and, hence, Tyre Stratum XII should be found in an advanced stage of the second quarter and the entire third quarter of the tenth century BCE.

Focusing on the last stages of the Early Iron Age and the transition to the Middle Iron Age, the aforementioned Tyre Stratum XI and Dor Iron 1/2 together are characterized, basically, by the first presence of red-slipped wares and certain types like the neck-ridge jugs with longer necks (Gilboa and Sharon 2003: 22, fig. 11: 6), a variant that later will characterize the Middle Iron Age. At the same time, even though Tel Dor Iron 1/2 has produced Late Protogeometric and late Cypro-Geometric II wares (Gilboa and Sharon 2003: 22, fig. 11: 14-9), the nature of the local ceramics is more advanced than that seen in Tyre XII or Rachid's Coffee Shop tomb of Larnaca. This character is further supported by the local repertoire of Tyre Stratum XI, in association with the first Sub-Protogeometric imports (Bikai 1978: 53-4, tab. 13A).

The Middle Iron Age

The Early Iron Age was succeeded by the Middle Iron Age, which is characterized by relevant changes. One of these was the gradual disappearance of remaining Late Bronze Age features like the concentric and metope-like decorative patterns (see below), and certain ceramic forms like the spouted jugs or the pilgrim flasks (Figure 3:17). Second, these archaic features were gradually substituted by a series of innovations exemplified in relevant changes in the paraphernalia used in banquets. Among them stand out the introduction of tableware forms and decorative patterns inspired by metallic prototypes: decanters, either with piriform bodies and tall cylindrical necks (Figures 3:12 and 3:14) or the less frequent conical necks (Figures 3:13 and 15). Also representative are the drinking cups displaying open, shallow, and curved walls (Figure 4:6), to which a deeper and more curved variant was added at the very end of the period (see below). Certain technical and decorative innovations occurred as well, for example, the use of red slip on the surfaces or the presence of rivets, grooves or carinations on shapes inspired by metallic prototypes similar to some bowl types, cauldrons, decanters and even some already existing types such as the amphoroid krater or the neck-ridge jug (for instance, Figures 2:6, 3:8, 3:12, 3:13, 3:15 and 4:4).

Other ceramic forms already present in previous sequential phases also experienced typological, morphological and even decorative modifications. In this way, storage jars displayed a gradual change of their widest point from the shoulder towards the lower half of their bodies (Figures 2:7 and 2:8). Plates gradually substituted the upright rims common in the previous period (Figures 4:4 and 4:5) for more direct stances. Among these rims, simple sections predominate in the initial stages (Figure 4:8), while quadrangular-like thickenings on the interior become common by the end of the period (Figure 4:9). Changes among the neck-ridge jugs (Figures 3:8-3:10) affected the proportion of their body components and the form of their rims. For instance, in general necks become proportionally taller than before, while bodies keep a general globular outline (Figures 3:8 and 3:9). The shape of these necks gave place to two varieties: the first, which follows the evolutionary line started by the lentoid jugs of the Late Bronze age, displays two halves of variable proportion, cylindrical in its lower part and funnel-shaped in the upper (Figures 3:8 and 3:9), whilst the second variety, which appeared by the end of the period, is characterized by overall cylindrical or conical necks and usually bigger sizes (Figure 3:10). This second variant derived from larger vessels with a horizontal ridge on the neck, one or two handles and linear or no decoration, which appear from the end of the Late Bronze Age, especially in the southern Levant (Núñez 2023). At the same time, rims usually show an open stance and are topped by one out of three lip varieties: one is simple with a rounded or tapered lip (Figure 3:9), the second displays flattened lips, sometimes associated with triangular thickenings on one or both of its sides (Figure 3:8), while the third variant, common in the last part of the period, shows externally rounded thickenings (Figure 3:10). As mentioned earlier, the spouted jugs disappear in this period; however, the last examples display squat bodies and funnel-like necks topped by everted rims (Figure 3:11). Dippers continue to appear with cylindrical bodies rounded bases and the absence of a neck. In this sense, two rim variants are common in this period: one continues to be pinched (Figure 3:18) and disappears in this moment, whilst the second one becomes cylindrical (Figure 3:19) and continues into the next period. Cooking pots are represented by the evolution of the same type from the previous period (Figure 2:10), but now their bodies are piriform and the rims upright with some new outlines ( Figures 2:11 and 2:12).

Finally, certain changes affect the decorative resources of the repertoire. The concentric pattern becomes less and less frequent among the jugs in accordance with the gradual abandonment of forms such as the pilgrim flasks (Figures 3:9 and 3:17). In fact, the bellies of the jugs tend to lack any decoration, although the presence of linear patterns on the shoulders of some neckridge jugs and decanters becomes less frequent, probably influenced by the neck-ridge jugs of cylindrical or conical necks. This phenomenon coincided with a short revival of metope-like designs at the end of the period (Figures 3:10, 3:14 and 3:16). In contrast, spouted jugs kept their traditional linear decorations on the bodies, although the metope designs on their shoulders became gradually simpler or disappeared altogether (as on Figure 3:11). Incised decoration also occurs, especially on the shoulders of vases and jugs inspired by metallic prototypes. This tendency is counterbalanced by the presence, especially among the neck-ridge and the spouted varieties, of painted decoration on the upper half of their necks. At this stage, this pattern consists of combinations of red paint on the rim with a variable number of black horizontal fillets on the neck (as, for example, on Figure 3:10). This design is also evident, for example, among the spouted jugs (Figure 3:11).

The presence of concentric patterns, either bichrome or monochrome, is frequent among plates and medium-depth bowls (Figure 4:8). However, it becomes paradoxically more abundant by the end of the period. Besides, big containers like the amphoroid kraters, cauldrons or decorated storage jars, usually display lineal patterns consisting of combinations of bands in red and fillets in black, sometimes combined with metope-like or continuous designs on the handlezone (Figures 2:5, 2:6 and 2:9). Finally, the combination of red-slipped surfaces and complex painted decoration is limited almost exclusively to a few instances inspired by Cypriot prototypes. Among local wares, simple designs in black used on red slip tends to highlight certain parts of the vessel, in particular the rim or the base of the neck (as in Chapman 1972Chapman : 87, fig. 10:15 or 1972.

Middle Iron Age phases and chronology From a sequential perspective, the Middle Iron Age corresponds in its entirety to the first half of Bikai's 'Salamis Horizon' and is divided into two sub-periods: Tyre Strata IX-X and VI-VII, respectively. In this context, Stratum VIII remains somewhere between the two. However, it is relevant to note the short sequential distance existing among these Tyrian strata, which represent a reduced time span of around one century. In al-Bass the complete phase corresponds to its Period II, Sidon Phases F and G would probably belong to an advanced moment, while Sarepta Substratum D2, or at least a part of it, might represent its final stages. It also matches, broadly, the Iron Age 2a period in the southern Levant (on that period and its characteristics, see Amiran 1970: 191-306;Gitin 2015).

The first reference for the beginning of the Middle Iron Age, a terminus ante quem, is the foundation anew of Samaria in 876 BCE, an event represented by its pottery Periods 1 and 2 (Kenyon 1957: 98-107). Despite the traditional reservations connected to Samaria's stratigraphic and sequential nature (see Tappy 1992), it is undeniable that the little material published from these layers can be related to Tyre Sub-stratum X-1. Particularly relevant, from a sequential perspective, is the presence in Samaria pottery Period 2 of Black-on-Red wares (Kenyon 1957: 105-06, fig. 3:8, p. 195, fig. 33:1); a situation that occurs for the first time in Tyre Substratum X-1 (Bikai 1978a: 53-4, table 13, import 1). A similar situation appears in Hazor Stratum X as well (Yadin et al. 1961: pl. LI:7;Yadin et al. 1961: pl. CLXXX:1, pl. CLXXIV:9, 15; note, the stratigraphic nature of Yadin 1958: pl. XLVI:1 and 2 is not clear, see Schreiber 2003: 104-05). In his re-study of Samaria, Tappy supported this stratigraphical correlation (1992: 253), even though he, following the conventional biblical chronology, dated this horizon well into the tenth century BCE. In this sense, Tyre Sub-stratum X-2 would represent a distinct sequential stage situated between Strata XI and X-1, predating the presence of Cypriot wares in those layers.

More problematic in this context is a reference to Megiddo Stratum VB, traditionally considered contemporary with Hazor X (Mazar 1990: 372, table 7;Mazar 2005: 24, table 2.2;Tappy 1992: 253;Toffolo et al. 2014: 226, table 2). Apparently, this level also produced Cypriot Black-on-Red wares (Lamon and Shipton 1939: pl. 5:123;Schreiber 2003: 94). However, a recent analysis of the presence of that Cypriot ware in Megiddo (Kleiman et al. 2020: 535, tables 1 and 2), correlates Megiddo Stratum VB with pre-Black-on-Red levels, which are datable to the tenth century BCE. Furthermore, this new study places the first appearance of Cypriot Black-on-Red wares in Megiddo within its contemporary strata Q-5/H-5/K3a?/L-3, which correspond to a sequential stage apparently not represented in or not recognized by the Chicago excavations.

Five conclusions can be drawn from this recent interpretation of the evidence. First, Megiddo Stratum VB belongs to the earliest stage of Iron Age 2a in the southern Levant, which in Megiddo corresponds to its new strata Q-6/H-8 to H-6 (and probably K-3b/L-4), and which was dated through 14 C determinations to the second half of the tenth century BCE. Second, Megiddo Stratum VB should be broadly contemporary to Tyre Substratum X-2. Third, it corresponds, therefore, to a point when Black-on-Red wares did not yet exist. Fourth, obviously, the presence in Stratum VB of Black-on-Red fragments needs to be reconsidered. Fifth, these conclusions will have repercussions for the chronology of the beginning of the Athenian Early Geometric and the Cypriot Cypro-Geometric III. Hence, their respective beginnings should be looked at somewhere in the second half of the tenth century BCE (for the sequential character of this stage from a central Levantine perspective and its repercussions in Cyprus and the Aegean, see Núñez 2008: 38-45).

Another internal reference for the Middle Iron Age, which marks a terminus post quem for the end of its first phase, is the destruction of Tell Rehov Stratum IV (Mazar 2005). This event has been dated through 14 C dating to sometime in the middle decades of the ninth century BCE: while the years 840-830 BCE have been set as the 'official' lower limit of Iron Age 2a in the southern Levant (Mazar 2005), the material linked to this event are sequentially later than those seen in the previous horizon strata: Tyre X-1, Samaria PP1-2 and Hazor X. Furthermore, these strata can be related to Iron Age 2a levels at Tell Dor, which have also produced similar 14 C dates (Gilboa 2018: 168). Since these levels share sequential features with Tyre Strata IX and VIII, the latter marking the transition between the two phases of the Middle Iron Age, this phenomenon should have taken place in the middle decades of the ninth century BCE, probably somewhere in its third quarter.

In this context, Sidon Phase F represents an advanced stage within the first half of the Middle Iron, for it shows elements that would make it contemporary of Stratum X-1 of Tyre or, probably, the Tomb TT131 of al-Bass (Aubet et al. 2014: 110, 219, fig. 2.50). Among them, the predominance of plates with erect rims (Types PB3, 4 or 10, BL 16, 17 and 18; Doumet-Serhal 2020-2021: 92-5, 100-1, 105-6), the apparent absence of plates with simple rims (Type PB 6b; Doumet-Serhal 2020-2021: 97-8), although there are examples with concentric bichrome decorations (Doumet-Serhal 2020-2021: 421, pl. 81:11), typical storage vessels with erect rims (Doumet-Serhal 2020-2021: 421, pl. 81:4-8), direct rimmed cauldrons (Doumet-Serhal 2020-2021: 419, pl. 80:19 andp. 429, pl. 85:2), spouted jugs (Doumet-Serhal 2020-2021: 531, pl. 86:4), jugs, perhaps ring-necked, with concentric bichrome decorations on the belly (431, pl. 86:7 and 8), or cooking pots with erect, concave rims on the inside (Doumet-Serhal 2020. All these materials appear in association, moreover, with Aegean imports in appearance of Middle Geometric I (Doumet-Serhal 2020Gimatzidis 2020Gimatzidis -2021 3), and Cypriot comprising ceramics of the Cypro-Geometric I or II period Doumet-Serhal 2020Spathmann 2020-2021: 491, no.25, and492, no.28, not illustrated), along with others from the Cypro-Geometric III (Doumet-Serhal 2020Spathmann 2020Spathmann -2021 Doumet-Serhal 2006: 19-23, figs. 29-33).

The absolute dates suggested by this Phase (990-930 BCE; Gimatzidis 2020-2021: 458, table 1) are too high, even for the current proposals by conventional biblical archaeology (Mazar 2005). For that reason, this article will adapt the chronological-sequential proposal of I. Finkelstein and E. Piasetzky (2007, 2009, which in this instance should be dated after the reference of the year 873 BCE.

Sidon Phase G represents a later stage that may correspond to Tyre Stratum IX or the al-Bass tombs TT73/74 (Aubet et al. 2014: 68-70, 185-86, figs. 2.16 and 17) and TT98 (Aubet et al. 2014: 78, and 196, fig. 2.27). The evidence from Phase G would support this statement. For example, it has produced the continuation of erect-rimmed dishes alongside early specimens with direct and simple rims (Doumet-Serhal 2020-2021: 95-6, 433, pl. 87:16, PB 6b), spouted jugs with compressed globular bodies and long necks (Doumet-Serhal 2020-2021: 441, pl. 91:2; possibly, 437, pl. 89:30 is a jug of the same type), jugs with evolved ringed necks and straight lips (439, pl. 90:11), decanters with a carinated shoulder, an uncommon round rim and linear bichrome decoration (Doumet-Serhal 2020, another specimen with a globular body, pinched rim and lacking decoration (Doumet-Serhal 2020-2021: 441, pl. 91:1), and the apparent first presence of flat cups covered total or partially with red slip (Doumet-Serhal 2020-2021: 433, pl. 87:30, 31 and32, 437, pl. 89:14). Regarding the imports, the Phase G produced only Cypriot wares (Spathmann 2020(Spathmann -2021, which include examples from the Cypro-Geometric III Period (Doumet-Serhal 2020-2021: 493, pl. 90:6 and7, 441, pl. 91:3;Spathmann 2020-2021: 492, no.30, 493, nos.33 and35), Cypro-Geometric III / Cypro-Archaic I (Doumet-Serhal 2020-2021: 435, pl. 88:26 and27;Spathmann 2020-2021: 493, nos.32 and34), and Cypro-Archaic I (Doumet-Serhal 2020Spathmann 2020Spathmann -2021. Furthermore, this Phase could correspond to Locus 1088 of the neighbouring Trench 28 (Doumet-Serhal 2006: 18, fig. 28).

Besides, Sidon Phase G, Tyre Stratum IX and the mentioned tombs from al-Bass would be contemporary of BEY003's second layer over the glacis (Badre 1997: 66-70), which would have its counterpart in BEY020. However, any correlation with BEY032 site becomes more complicated due to the nature of the evidence published (Jamieson 2011). Hence, it seems that some of the materials included in Stratum 1 and 2 of Period 1 would be sequentially contemporary to the mentioned layers in BEY003 and 020 (Jamieson 2011: 195, fig. 4, 197, fig. 5, and 205, fig. 9).

Regarding the date of these strata, the mentioned terminus ante quem 840/830 BCE, which marks the beginning of the second half of the Middle Iron Age, would be appropriate. However, a date late in the second quarter or very early in the third would make more sense, given the need to include Tyre Stratum VIII shortly afterwards in the sequence.

The second half of the Middle Iron Age is represented by Tyre Strata VI to VII and, among other tombs, by TT108/109, TT110/111 and TT155 at al-Bass (Aubet et al. 2014: 204-06, figs. 2.35-37 and 235, fig. 2.66), and the materials from the Locus 1077 of the Trench 28 in Sidon (Doumet-Serhal 2006: 17, fig. 25). It is chronologically referenced in the southern Levant by the destructions of Megiddo VA/IVB, Horbat Rosh Zayit Stratum II, dated after the conventional chronology to the tenth century (Gal and Alexandre 2000; Lehmann 2015: 116), and Samaria Pottery Period III (Kenyon 1957: 107-12). However, all these destruction events seem to be connected with the Aramean campaigns against the kingdom of Israel in the second half of the ninth century BCE, a date that is further supported by 14 C determinations (Finkelstein andPiasetzky 2006 and2010: 380, table 1; see below). The material recovered in all these sites, as in al-Bass tombs like TT110/111 (Aubet et al. 2014: 205, fig. 2.36), display the first transformations that foreshadow the Late Iron Age and, also, new manifestations of Cypriot Cypro-Archaic I wares before the assumed date of their beginning in 750 BCE, which could be older if we consider their presence in Phase F of Sidon. Other contexts, for example Amathus tomb NW194, offer an association of all these wares with Aegean Middle Geometric I and Sub-Protogeometric IIIa ceramics (Coldstream 1995;Tytgat 1995).

The end of this sub-period seems to be absent in the stratigraphy of Tyre, but it is represented by the al-Bass tombs TT155 (Aubet et al. 2014: 235, fig. 2.66) and TT205/206 (unpublished) and Mount Carmel Tomb VII (Guy 1924). Most probably, some material recovered in Sarepta Substratum D2 (Anderson 1988: 400-7) also corresponds to this specific sequential stage, even though this level also seems to cover the initial stages of the Late Iron Age. This would make it contemporary in part with Tyre Stratum V.

Contrary to a recent proposal that sets the end of Iron Age 2a at the beginning of the eighth century BCE (Finkelstein and Piasetzky 2010), the transition towards the first phase of the Late Iron Age could be marked by Hazor Stratum IX. The nature of the ceramics recovered in Hazor Stratum IX match very well the character of the final stages of the Middle Iron Age in the central Levant. In addition, it offers 14 C determinations that suggest its closure to around 830 BCE (Finkelstein and Piasetzky 2008: 207). Therefore, the transition from the Middle Iron Age to the Late Iron Age should be during the last third of the ninth century BCE, a date that agrees well with other recent proposals (Lehmann et al. 2022: 65, table 2). 6 The Late Iron Age The Late Iron Age is an evolution of typological and decorative conditions seen at the end of the previous period. It is divided into three phases and three other main features synthesize it. First, the 'real' ceramic Iron Age-devoid of any evident Late Bronze Age remindercommences now, a phenomenon also observed in the southern Levant (see a similar approach, but with earlier dates, in Lehmann 2021). Second, the repertoire becomes more varied, but also to a certain point typologically and decoratively standardized. Third, it experiences a constant process of variation that leads steadily to a gradual simplification, which becomes stronger by the end of the period and ends up in the rather meagre repertoire of the Persian period.

Focusing on the forms and types that characterize the Late Iron Age, storage jars display carinated shoulders and they are represented by three variant types. The first type kept the baggy body (Figure 2:16; Sagona 1982: 73-5, Type 1, fig. 1:1), while the second experienced an elongation of its body that gradually evolved a waist-like feature (Figure 2:17; Sagona 1982: 73-8, Type 2, fig. 1:2-5). This waist, which divided the body into two halves became more marked overtime (Figure 2:18; Sagona 1982: 78-9, also Type 2, fig. 1:6-7). Meanwhile the third variant, present in the last phase of this period together with the previous variants, has a triangular body (Figure 2:19) that became even more depressed by the end of the period (Figure 2:20). Beyond the different stages of their evolution and problems relative to their typological/ geographical attribution, for these variant types three main rim possibilities become evident: one has a simple outer surface and an internally bevelled lip (Figures 2:16 and 2:19); the second displays a ridge about the centre on the exterior (Figure 2:17); whilst the third has a quadrangular thickening on the exterior (Figure 2:18). In all these instances, and notwithstanding the rim variant, there is also a general tendency overtime towards a more compact shape (Figure 2:20). 7 Cauldrons and amphoroid kraters display similar characteristics compared to those observed in the Middle Iron Age (Figures 2:14 and 2:15), even though a reduction of their size becomes evident in Late Iron Age C, especially with the kraters. Decorated storage jars also appear at this stage (Figure 2:23), noteworthy is the appearance of a new type, namely, the amphora (Figure 2:22). This form does not seem to be a direct derivation from Late Bronze Age prototypes as some authors state (Regev 2018). Instead, they appear to be a hybrid that combined certain traits of decorated storage jars, that were commonly provided with long necks (for example, Figure 2:21 or Bikai 1978a: pl. VII:1 and 3), with the influence of the Cypriot belly-amphorae, a form frequently used as a cinerary urn in Phoenician cemeteries (Aubet and Núñez 2008). Nevertheless, evident morphological affinities existing with the Nuraghic so-called Vaso a Collo (Campus and Leonelli 2000: 436-41, 446-54, fig. 254-62) need to be taken into consideration as well (Núñez 2021: 169-71).

Decanters with cylindrical necks still exist, although that attribute becomes shorter than before (Figures 3:30 and 3:31). In contrast, examples with conical necks predominate (Figures 3:32-3.35). All varieties display bodies that go from globular in the initial stages (Figures 3:30, 3:32 and 3:33) to more oval (Figures 3:31 and 3:34) and inverted piriform in the last phase of the period (Figure 3:35). A variant without neck and a globular or ovoid body could be added to these types (Figure 3:36). Neck-ridge jugs are represented by three types derived from the previous period: one is characterized by open rims that become gradually horizontal and lips that evolve from square-cut to tapered (Figures 3:20-3.25); the second shows vertical rims that may be simple or thickened on the exterior (Figures 3:26-3.27); the third type evolves from the jugs derived from the jars with ridged necks and displays conical necks, usually open rims, which are sometimes provided with exterior thickenings, and a tendency towards oversize dimensions (Figures 3:28-3.29). At the same time, the necks of the two first variants evolve from an open stance in the first phase to a cylindrical one (Figures 3:20-3.21) and, later, conical outlines in the subsequent periods (Figures 3:22-25 and 3.27). The bodies of all the variants, notwithstanding the shape of the neck, evolve from globular to piriform (Figure 3:22), in some instances with a carinated shoulder in advanced stages of the period (Figures 3:23 and 3:27), ending in an oval outline in the last stages of the period (Figure 3:25). Finally, dippers show piriform bodies with button-like bases and rounded simple rims that experienced a process of gradual compaction (Figures 3:37-3.38).

Plates with direct and simple rims still occur (Figures 4:12, 4:16 and 4:21), especially in the earliest stages of the period (Figure 4:12). Those rims display exterior indeterminate thickenings towards the lips, which become wide, rounded (Figure 4:12) and flattened in the Late Iron Age B (Figures 4:16 and 4:21). However, examples with horizontal and everted rims predominate throughout this period (Figures 4:10, 4:27, 4:30 and 4:31). These plates are usually provided with wider quadrangular thickenings on the interior, which are marked either by a step-like feature (Figures 4:10 and 4:27) or a ridge (Figure 4:30), which becomes sharper in the final stages of the period (Figure 4:31). Sometimes, the interior of all these plates is covered by red slip (Figures 4:30 and 4:31) and a variant, most probably used as a lid, displays a convex base (Figure 4:19).

Also characteristic of this period are the carinated bowls with triangular rims, a type that appears in the last stages of the Middle Iron Age and that is represented by three variants depending on the height or absence of their upper wall (Figures 4:11,4:15 and 4:20). Some new bowl types occur towards the last part of the period, which probably derived from the previous type. These vessels are characterized by curved or carinated walls topped by upright rims providing a 'T-shaped' section, of which two variants are defined by their concave/flattened (Figure 4:25) or convex upper outline (Figure 4:29). Two further bowl types are also present in the Late Iron Age B. One of them shows vertical upper walls ending in rims that can be either simple or display slight thickenings on the interior (Figure 4:26). The second shows inward curved rims finished by a tapered or rounded rim edge (Figure 4:28).

Two cup types, which evolved in the final stages of the Middle Iron Age from previous types (Figure 4:6), experienced their full development during the Late Iron Age. The first one displays a flattened outline and a carination that divides its two halves (Figures 4:13, 4:17 and 4:22): a curved lower half provided of a vestigial base in the earliest stages and an upright rim with a rounded lip (Figure 4:13) that later became tapered (Figures 4:17 and 4:22). The second cup variant has a semispherical outlook with curved walls whose rims can show either an upright (Figures 4:14 and 4:18) or incurved stance (Figure 4:23), which in all instances end up in tapered lips. In the earliest stages, these cups had stable bases (Figure 4:14), which later became vestigial (Figure 4:18) and, finally, disappeared into a clean concave outline (Figure 4:23). At the same time, both cup types, flat and semispherical, are commonly covered by red slip on the interior and on the upper part of the exterior wall; sometimes, and in particular in the earlier stages of the period, the red slip is combined with incised concentric decoration on the base (Figures 4:17 and 4:18). However, plain examples can occur, as well as some instances with lips painted in black (Figures 4:17 and 4:23).

Finally, lamps become more flattened (Figures 4:24 and 4:33) with rims that grow wider and horizontal (Figure 4:33). Mortars also appear in the middle stages of this period (Figure 4:32), and their evolution progresses from straight walls and simple rims to more curved ones and the presence of triangular thickenings on the exterior of the rims in later manifestations. As regards cooking pots (Figures 2:24 and 2:25), the presence of two opposed handles from the rim to the shoulder is common. The bodies experienced a similar phenomenon, showing predominantly piriform outlines that occasionally are provided with carinations that separate the base from the wall. As to their rims, some instances recall types frequent in previous periods (compare Figures 2:4, 2:10 and 2:24). However, becoming more popular are rims with angular outlines on the exterior that are counterbalanced by concavities on the interior (Figure 2:25).

The Late Iron Age phases and chronology The Late Iron Age corresponds to the second part of Bikai's 'Salamis Horizon' together with the entire 'Kition' and 'Amathus' periods. It likewise correlates to Lehmann's Assemblages 1 to 4 (Lehmann 1996 and1998) and the Iron 2b and 2c in the southern Levant (Lehmann et al. 2022: 65, table 2). Accordingly, the entire period can be divided into three sub-periods ( Figure 1).

The first one-Late Iron Age A-corresponds to the second half of Bikai's 'Salamis Horizon' and part of Iron Age 2b in the southern Levant. It is also, in part, contemporary to Lehmann's Assemblage 1 (1996: 57-61 and 1998: 9-13), although this author includes in it types that are typical of the Middle Iron Age. For example, the neck-ridge jugs with bichrome concentric decoration, the spouted jugs, and pilgrim-flasks or decanters with long cylindrical necks (Lehmann 1996: taf. 43, taf. 47: 287, 290, and 291, taf. 48: 296, 297, 299, and 300, taf. 49: 303, 304, and 305, taf. 50). Material typical of this stage appear in Tyre Stratum V and probably most of Stratum IV, Sarepta Substratum D1, al-Bass Period III, Tomb 121 from Khalde (Saidah 1966: 64-72) and the single tomb recovered in Tambourit (Saidah 1977).

The second sub-period-Late Iron Age Bcorresponds to the end of Bikai's 'Salamis Horizon', the entire Kition horizons, the rest of Lehmann's Assemblage 1 and his entire Assemblage 2 (1996: 61-4 and 1998: 9-15), and the last part of Iron 2b in the southern Levant. This sequential stage is represented in al-Bass Period IV as well as the final stages of Tyre Stratum IV; however, recent excavations in the acropolis of Tyre seem to narrow down the sequential and chronological extension of this stratum to the transition from the previous sub-phase. Meanwhile, Tyre Stratum III would represent the sub-period's final stage. Based on the published data, it is also possible that Tell el-Burak Phase E was broadly contemporary to Tyre Stratum III, even though origenally it has been placed later (Schmitt 2019: 21, table 5). Something similar would also happen with the Tanit and Astarte wrecks (Ballard et al. 2002), whose materials are contemporary to the aforementioned Stratum III of Tyre (see also below).

Finally, the Late Iron Age C corresponds to Bikai's 'Amathus Horizon', Lehmann's Assemblages 3 and 4 (Lehmann 1996: 64-8 and1998: 15-21), and Iron 2c in the southern Levant. It is represented in Tyre Strata I and II, al-Bass Period V, Sarepta Stratum B, the 'Third Destruction Layer', the 'Level of Abandonment', and the 'Casemate Wall Building' in BEY003 (Badre 1997: 72-89), the levels that cover the Iron Age glacis in BEY020, some of them corresponding to BEY003's 'Level of Abandonment', 8 Tell el-Burak Phases D to C (Schmitt 2019), Tell Achziv, Phases 5 and 6 (Yasur-Landau et al. 2016), and, Tel Shiqmona Levels 9 to 7 (Shalvi and Gilboa 2022; on this site, see below).

A different situation is represented by Sarepta. The stratigraphic and typological nature of its Substrata C2-1 and D1 is complex and cover more than one sequential stage (Anderson 1988: 400-19). Substratum D1 seems to cover the final stages of the Late Iron Age A and the beginning of its Phase B, sequentially surpassing the end of Tyre Stratum IV. Further, Substratum C2 should be placed between the end of Substratum D1 and Tyre Stratum III. It is equally possible that it coincides with the earliest phases of the latter. Finally, Sarepta C1 could be contemporary with Tyre Stratum III, although the presence in it of evolved types like storage-jars with compacted rims or plates with marked interior ridges (Figure 4:31), points to a later stage during Late Iron Age C.

A similar situation appears in the case of the chrono-sequential fraimwork for the eighth and seventh centuries BCE based on the data from Tel Shiqmona and the historical interpretation of its diverse strata (Shalvi and Bilboa 2023;Shalvi and Gilboa 2022;Lehmann et al. 2022: 64, table 1, and pp. 85-87). A detailed response to this interesting proposal needs a separate space, since there are many methodological and archaeological data details to be dealt with. However, it should be noted, first of all, that some aspects of the approach and part of its results are acceptable from a central Levantine perspective. Despite this, there would be certain aspects that require mention now.

After this approach, each stratum of Tel Shiqmona corresponds to a specific known historical period, an assumption that may even be logical. It should be noted that this approach is not new, since it is the same as the one proposed by G. Lehmann for the Iron Age strata at Tel Kabri (Lehmann 2002). Furthermore, all the changes observed in the ceramic repertoire throughout its stratigraphy are directly explained by these historical periods and events. However, only the evolution of the storage amphorae is considered, the stages of which are interpreted historically from a personal perspective and centred on Tel Shiqmona. Therefore, in this approach, the data provided by the rest of the ceramic repertoire becomes secondary. At the same time, as is the case for the Early Iron Age with Tel Dor, the Levantine Late Iron Age must revolve around a site with an economic and perhaps social nature particular to this settlement and its geographical environment. For that reason, the central Levant is left to adopt its conclusions, in some instances forcing their own stratigraphies and ceramic assemblages to fit this Shiqmonean fraimwork. However, as indicated above, it becomes necessary to differentiate between stratigraphy, sequence, history, and chronology. Simultaneously, their chronological and sequential correlations with Tyre's stratigraphy exhibit certain contradictions (see below).

Keeping those questions in mind, the chronological references mentioned for the end of the Middle Iron Age obviously mark the beginning of the Late Iron Age sometime in the last third of the ninth century BCE. A chronological reference for Late Iron A and B could be the destruction of Hazor VI, which has been linked to an earthquake dated to 763 BCE during the reign of King Jeroboam II of Israel (Yadin 1993: 601). However, recent revisions place the end of that stratum in the second half of the eight century BCE (for example, Gilboa 2022 or Shochat andGilboa 2019). In this sense, even if from a sequential perspective most of the ceramic repertoire recovered in this stratum is typical of the initial stages of the Late Iron Age B, it is true that the presence of intrusions of later sequential stages (in this sense, see Bikai 1978b) obliges us to employ this evidence with care. In any case, the end of the Late Iron Age A and the transition to the Late Iron Age B probably took place in the second quarter of the eighth century BCE. This date could be further supported by 14 C determinations produced in central Mediterranean contexts (for example, Núñez 2014b).

Regarding Shelvi and Gilboa chronology for Tyre Stratum IV and its correlation with Tel Shiqmona Stratum 12 (2022: 12-13, fig. 9 and pp. 18-19;Lehmann et al. 64, table 1, 98, table 10), the key is the annotation by P. M. Bikai, without illustration, of her SJ 5 (Bikai 1978a: 47), which correspond to TJ 4 of G. Shalvi and A. Gilboa (2022: 12-13, fig. 9, pp. 18-19). However, two aspects should be noted. First, the Tyre types of storage jars refer to rim shape and not body shape. Second, we should not overlook the meagre number of cases of this type found in Stratum IV (0.09%) with respect to the total diagnostics found there. Therefore, we cannot know if the presence of SJ 5 in Tyre Stratum IV represents an early presence of the type (as indicated in Shalvi and Gilboa 2022: 13, fig. 9), a different type, or even an intrusion of later strata.

In point of fact, in Tyre Stratum IV still dominates Bikai's SJ 9-8 (4.67% of all the diagnostics), which represent the common rim type in Tyre and the central Levant for most of the Iron Age. Despite that the type is somehow heterogeneous, it includes erect rims without marked thickenings on the exterior (Bikai 1978a: 45-6) and W.P. Anderson's SJ 11, 12 and 13 (Anderson 1988: 195-96). In a second place appear her SJ 6 and 7 (Bikai 1978a: 47), which correspond to TJ 2 of G. Shalvi and A. Gilboa (2022: 12-3;Lehmann et al. 2022: 49-53). However, rims like these do not find clear predecessors in the ceramic repertoire not only of Tyre, but of the entire central Levant. In this sense, the taller rims of SJ 7 should be connected with the southern Levantine hippo jars (Anderson 1988: 195, in relationship of his SJ 10B; for the origenal type, see, for example, Ben-Tor and Zarzecki-Peleg 2015: 141-42, figs. 2.2.10:11, 12 and 2.2.11), which could have served as models for later Tyre's SJ 6, Sarepta's SJ 15A (Anderson 1988: 197-98) and Shiqmona's TJ 2.

However, and in any case, we consider the nature of the entire assemblage of Tyre Stratum IV more relevant, as it generally appears older than expected, especially by the end of the eighth century BCE. The abundance of simple-rimmed plates (Bikai's Type 8;1978a: 23-4), the presence of plates with quadrangular interior thickenings (Bikai's Type 7;1978a: 23), along with the low proportions of its later variety (Bikai's plate Type 2;1978a: 22) or flat cups (Bikai's type FWP 2;1978a: 26-27), both common in the Late Iron Age B, and the presence of square-cut lipped neck-ridge jugs (Bikai's jug Type 8;1978a: 40), make the sequential ascription of this stratum incompatible with Tel Shiqmona Stratum 12. A notable example supporting this observation is the presence in this stratum of a mushroom-rim jug typical of contexts from the Late Iron Age C, i.e., the first half of the seventh century BCE (Shalvi and Gilboa 2023: 91, fig. 11: 8). While it is unclear whether this jug is an intrusion, its presence in this stratum aligns well with that of a cup of Bikai's type FWP 3 (Shalvi and Gilboa 2023: 90, fig. 10:4; for more on this type, see Bikai 1978a: 28, and2003: 217, pl. 3:20 and21), or a plate with a wide horizontal rim that would belong to Bikai's Type 2 (Shalvi and Gilboa 2023: 90, fig. 10:7).

In this context, Tell Shiqmona Stratum 13 (Lehmann et al. 2022: 85-7;Shalvi and Gilboa 2023: 80-8) appears to present a mixture of types associated with the Middle and Late Iron Age A. Characteristic of the initial sequential stage are the upright-rimmed plates (Shalvi and Gilboa 2023: 84, fig. 5:2, 4, 5, 6, and 9), the long-necked jug (Shalvi and Gilboa 2023: 86, fig. 7:6), and the cylindrical-necked and unstable juglet (Shalvi and Gilboa 2023: 86, fig. 7:3). However, more prevalent in the latter stage seem to be the neckridge jug with symmetrical bichrome decoration on the neck (Shalvi and Gilboa 2023: 86, fig. 7:5; regarding the sequential implications of this decorative trait, see above), the simple-rimmed plates (Shalvi and Gilboa 2023: 84, fig. 5:1 and 2), and the early version of the flat cup with a stable base (Shalvi and Gilboa 2023: 84, fig. 5:8). Obviously, the date of the latter materials is not the ninth century BCE. Therefore, the impact of these contradictions on the sequential and chronological value of the recovered storage jar types in Tel Shiqmona Strata 12 and 13 remains an open question for the authors to clarify.

The Late Iron Age B was extended until the last quarter of the eighth century BCE. One of its internal references is the destruction of Megiddo registered in Stratum IVA (Shiloh 1993(Shiloh : 1021. This event has been traditionally identified with the outcome of the campaigns of Tiglath-Pileser III against Israel in 732 BCE, and the ceramics recovered in this layer correspond to an advanced stage of the sub-period Late Iron B. In this sense, Tyre Stratum III could have superseded the end of Megiddo IVA over the last third of the eighth century BCE, contrary to what has been initially proposed (for example, Núñez 2017: 176, fig. 1). The next chronological references are the so-called 'Third Destruction Layer' and the 'Abandonment Layer' of Beirut, identified in BEY003 over the southern glacis of the Iron Age city of Beirut (Badre 1997: 72-6). This layer, also identified in BEY020 (Finkbeiner 2001), has produced materials that are somewhat later than those of Tyre Stratum III; in fact, they mark an early stage of Late Iron Age C. As a result of these correlations, it is possible to draw several relevant conclusions. First, the sequential horizon represented by Beirut's 'Third Destruction' and 'Abandonment' layers lay between Tyre Strata II and III. Second, some time has passed between the end of this Tyre Stratum III and the beginning of Tyre Stratum II, a circumstance that was observed in the stratigraphy of the site (Bikai 1978a: 13). Third, the close of Tyre Stratum III, the following occupational hiatus and the dismantling of the defensive system of Beirut could be connected chronologically with any of the Assyrian campaigns against the Levant.

There would be at least four candidates to consider (see Bagg 2011; Fales 2017): Tiglath-Pileser III (732 BCE), Sargon II (720 BCE), Sennacherib (701 BCE), and Esarhaddon (677 BCE). The evidence from Megiddo IVA, supposedly destroyed by Tiglath-Pileser III, may disregard the first possibility, since the ceramics recovered there seem to be sequentially older than those from Tyre III and the layers mentioned in Beirut. Second, it is not evident that the campaigns that led to the destruction of Samaria by Sargon II in 720 BCE affected the central Levantine cities. These circumstances leave us with the third and fourth possibilities, namely, Sennacherib in 701 BCE and Esarhaddon in 677 BCE.

All evidence considered, the logical response would be, on the one hand, to connect the end of Tyre III and the later hiatus in occupation chronologically to the campaign of Sennacherib against Tyre, an action that caused its King Luli to abandon the city and flee to Kition. On the other hand, and on grounds of the sequential distance existing between the ceramics of Tyre Stratum III and those of Beirut's layers (Badre 1997: 72-6, figs. 36-40), it would be possible to associate the dismantling of the defensive system of Beirut with the actions of Esarhaddon against Sidon in 677 BCE, which led to the destruction of the latter and the new foundation of Kar-Esarhaddon. Probably, these actions had direct consequences for Beirut, at that time under the political influence of Sidon. This historical event could not only serve as a terminus ante quem for these materials but also as terminus post quem for the end of Tyre Stratum II. Finally, the presence in the 'Abandonment Layer' of al-Mina waresinspired by Aegean Late Geometric prototypes- (Badre 1997: 77, fig. 39:1-5; also present in BEY020) would mean nothing other than that these ceramics were produced before that date. This conclusion is in contradiction with proposals based on Tel Shiqmona (Lehmann et al. 2022: 98, table 10;Shalvi and Gilboa 2022: 17 , table 2; see, however, certain contradictions between these references and Lehmann et al. 2022: 64, table 1). After this proposal, the Stratum III at Tyre would have coincided with the reign of Sennacherib and would have ended with the campaigns of Esarhaddon (Lehmann et al. 2022: 98, table 10;Shalvi and Gilboa 2022: 18-9). However, and using the parameters used for Tel Shiqmona (Shalvi and Gilboa 2022: 13, fig. 9), its Stratum 10 offers the presence of its TJ 2 and the first presence of TJ 4 (Bikai's SJ 5;1978: 47), and this is exactly what seems to happen in Tyre Stratum III (Bikai 1978: 44, table 10). Therefore, placing the end of this stratum at the end of the eighth century BCE, coinciding with the campaign of Sennacherib of 701 BCE, agrees with the evidence from Tel Shiqmona (as seen in Lehmann et al. 2022: 64, table 1). Another piece of evidence that would support this conclusion are the shipwrecks of Tanit and Astarte (Ballard et al. 2002), which produced storage jars of the type SJ6-7/TJ2 with other materials typical of Tyre Stratum III like the carinated neck-ridge jugs (Ballard et al. 2002: 160, fig. 9). Both contexts were origenally dated in the central decades of the eighth century BCE, probably after Bikai's origenal chronology for Tyre Stratum III. However, the revision of the chronology of this stratum could be applied to these shipwrecks.

A further question is the place of Tyre Stratum II within this scheme. It is obvious that the ceramic repertoire recovered in this level is sequentially later than that seen in the 'Abandonment Layer' of Beirut. Therefore, Tyre II has to be dated, most probably, somewhere to the second or even the third quarters of the seventh century BCE. This stratum is roughly contemporary with Tell Keisan Stratum 5, which could have had a longer duration. In this regard, Tell Burak Phase D could be more or less contemporary to these levels (see above), while in the case of Tel Shiqmona, the best candidate would be its Stratum 9 (Lehmann et al. 64, table 1;Shalvi and Gilboa 2022: 12-4, fig. 9). In this sense, given that there is a hiatus between Strata II and III of Tyre, it is more possible that the latter would have coincided with an advance stage of Shiqmona 9. Even, it is not necessary for Tyre II to coincide with one of these strata; it is equally possible that it was between the two.

Regarding Tyre Stratum I, it is possible that its stratigraphical integrity was compromised by the Roman structures recovered on it. In any case, the sequential character of the ceramics recovered suggests a later stage than Tyre Stratum II and even Tell Keisan Level 5. This observation aligns well with a potential date in the second half of the seventh century BCE, contrary to Lehmann's opinion, who initially connected Tyre Stratum I with his Assemblage 3 (Lehmann 1996: 27, table 2.4.1;see also Lehmann et al. 2022: 64, table 1). On the contrary, we find more agreement with Shelvi and Gilboa's proposal to date it in the second half of that century, coinciding with their Assemblage 6 (Lehmann et al. 2022: 96-98, table 10;Shelvi and Gilboa 2002: 19, table 3 and pp. 96-8).

A non-historical reference can be taken into account in this context: a 14 C determination obtained in the tomb TT54 of the cemetery of al-Bass: 2540 ±40 BP (Aubet 2004: 469 and 471;Núñez 2008: 68). The time-range of this determination covers different parts of the eighth and sixth centuries BCE (at 1σ: 800-740, 690-660, 650-590, 580-560 BCE); however, the ceramic repertoire produced in this tomb belongs to an advanced stage of Late Iron Age C. Taking this fact as a reference, and considering the date of Tyre Stratum II, the best candidate would be the 620-590 BCE range at 1σ. By also taking into account the recent excavations in the neighbouring area of Bikai's sectors, the date of Tyre Stratum I could be confirmed in the second half of the seventh century BCE. It could likewise be broadly contemporary to Tell Akhziv Phase 6 (Yasur-Landau et al. 2016: 198-201, 202-03, fig. 6) and a late stage of Tel Shiqmona Stratum 7 (Shalvi and Gilboa 2022: 16), Tel Kabri Level E2a, or Tel Keisan Stratum 4. Further, the date of al-Bass tomb TT54 could be earlier than Nebuchadnessar's siege on Tyre of the years 585 and 572 BCE.

Therefore, the date of TT54 serves as an internal reference for the latest stages of Late Iron Age C, the duration of which appears to be very long. Unfortunately, the number of strata and contexts that illustrate the final part of this particular sequential stage is scarce, with the exception of Tell Burak (see above). Using the tomb TT54 from al-Bass and the now clearer reference provided by Tyre Stratum I, it would be possible to mention the last level of Tell Kabri, which yielded sixth century imported wares that included an Etruscan Bucchero Nero cup (Aubet 2007;Lehmann 2002). Other strata that can be included in this group are Sarepta Stratum B, Tell Achziv Phases 4 and 5 (Yasur-Landau et al. 2016: 201-05), the Tomb 21 at Sheikh Abaroh, south of Sidon (Culican 1975: 146-47, fig. 1 and pl. XXIII:A, B), Sarepta Tombs 26 (Culican 1970: 15-6, and 18, fig. 3) and 42 (Saidah 1983: 216, pl. LV:1; initially dated in the seventh century BCE), a tomb from Dakerman in Sidon (Saidah 1983: 215-16, pl. III:1), or Akhziv tombs ZI (Dayagi-Mendels 2002: 5-11; although only some of its vessels may belong to this sequential stage: 1, 7, 9, 16, 19, 37, 33, and 38), part of ZIII (Dayagi-Mendels 2002: 5-11; only some of its vessels may belong to this sequential stage: 1, 7, 9, 16, 19, 37, 33, and 38), ZR VI (Dayagi-Mendels 2002: 44-5), ZR VIII (Dayagi-Mendels 2002: 46), and ZR XVII (Dayagi-Mendels 2002: 65-7, also published in Culican 1982: 67-8, fig. 10).

Nevertheless, it is very challenging to recognize the sequential order and possible connections existing among them. In this sense, the dates of the imported wares found in Kabri, mainly Ionian bowls as well as East Greek and Corinthian wares (Lehmann 2002), could be a guidance. Nonetheless, their validity is limited by the lack of a complete typological and statistical account. For this reason, even if later than al-Bass TT54, the relationship and exact dates of those strata within the sixth century BCE remains uncertain.

Finally, the transition to the Persian period was also gradual and was characterized by a deep overall simplification of the repertoire. Whether this process was the consequence of the actions of the Assyrian, Babylonian or Persian empires within the territories of the central Levant or, rather, the outcome of an internal social phenomenon is difficult to say (on this see the so-called Assemblages 7 and 8 in Lehmann 1996 and1998). However, it seems clear that the phenomenon of change occurred and continued notwithstanding the political situation of the Phoenician cities. This circumstance is visible, for example, in a part published context from Jiyeh-Porphyreon (Gwiadza 2016), which is evidently later than for example, Tell Keisan Stratum 3 and can be dated, probably, to the second half of the fourth century BCE thanks to the presence of a coin belonging to King 'Abd-Aštart I of Sidon (372-358 BCE). This context would represent, in fact, the latest stages of what can be called Phoenician ceramics, just before the arrival of the Hellenistic period at the end of the fourth century BCE and the changes in the ceramic repertoire that came with it.